We respect the reproduction of our works and believe it will help to motivate us to run faster. Hence, we are gradually releasing our simulation codes on Github.
All manuscripts within my website are subject to the copyright law. Paper downloading is limited to academic research only. For more information, read the Copyright Notice.
1109573
XMUWZA6J
items
1
ieee-transactions-on-communications
0
date
desc
year
1
title
R. Li
18
https://www.rongpeng.info/li/wp-content/plugins/zotpress/
%7B%22status%22%3A%22success%22%2C%22updateneeded%22%3Afalse%2C%22instance%22%3A%22zotpress-6354d89b8eaf2b6d9fb81ca3a6ccf172%22%2C%22meta%22%3A%7B%22request_last%22%3A100%2C%22request_next%22%3A50%2C%22used_cache%22%3Atrue%7D%2C%22data%22%3A%5B%7B%22key%22%3A%22GA2ZZSDY%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Lu%20et%20al.%22%2C%22parsedDate%22%3A%222024-03%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EZ.%20Lu%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20X.%20Chen%2C%20E.%20Hossain%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%5C%22%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_COMST2024.pdf%27%3ESemantics-empowered%20communication%3A%20A%20tutorial-cum-survey%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EIEEE%20Commun.%20Surveys%20Tuts.%3C%5C%2Fi%3E%2C%20vol.%2026%2C%20no.%201%2C%20pp.%2041%5Cu201379%2C%20Mar.%202024%2C%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DGA2ZZSDY%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Semantics-empowered%20communication%3A%20A%20tutorial-cum-survey%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhilin%22%2C%22lastName%22%3A%22Lu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xianfu%22%2C%22lastName%22%3A%22Chen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Ekram%22%2C%22lastName%22%3A%22Hossain%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22Along%20with%20the%20springing%20up%20of%20the%20semantics-empowered%20communication%20%28SemCom%29%20research%2C%20it%20is%20now%20witnessing%20an%20unprecedentedly%20growing%20interest%20towards%20a%20wide%20range%20of%20aspects%20%28e.g.%2C%20theories%2C%20applications%2C%20metrics%20and%20implementations%29%20in%20both%20academia%20and%20industry.%20In%20this%20work%2C%20we%20primarily%20aim%20to%20provide%20a%20comprehensive%20survey%20on%20both%20the%20background%20and%20research%20taxonomy%2C%20as%20well%20as%20a%20detailed%20technical%20tutorial.%20Specifically%2C%20we%20start%20by%20reviewing%20the%20literature%20and%20answering%20the%20%5Cu201cwhat%5Cu201d%20and%20%5Cu201cwhy%5Cu201d%20questions%20in%20semantic%20transmissions.%20Afterwards%2C%20we%20present%20the%20ecosystems%20of%20SemCom%2C%20including%20history%2C%20theories%2C%20metrics%2C%20datasets%20and%20toolkits%2C%20on%20top%20of%20which%20the%20taxonomy%20for%20research%20directions%20is%20presented.%20Furthermore%2C%20we%20propose%20to%20categorize%20the%20critical%20enabling%20techniques%20by%20explicit%20and%20implicit%20reasoning-based%20methods%2C%20and%20elaborate%20on%20how%20they%20evolve%20and%20contribute%20to%20modern%20content%20%26%20channel%20semantics-empowered%20communications.%20Besides%20reviewing%20and%20summarizing%20the%20latest%20efforts%20in%20SemCom%2C%20we%20discuss%20the%20relations%20with%20other%20communication%20levels%20%28e.g.%2C%20conventional%20communications%29%20from%20a%20holistic%20and%20unified%20viewpoint.%20Subsequently%2C%20in%20order%20to%20facilitate%20future%20developments%20and%20industrial%20applications%2C%20we%20also%20highlight%20advanced%20practical%20techniques%20for%20boosting%20semantic%20accuracy%2C%20robustness%2C%20and%20large-scale%20scalability%2C%20just%20to%20mention%20a%20few.%20Finally%2C%20we%20discuss%20the%20technical%20challenges%20that%20shed%20light%20on%20future%20research%20opportunities.%22%2C%22date%22%3A%222024-3%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_COMST2024.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222024-03-09T07%3A23%3A27Z%22%7D%7D%2C%7B%22key%22%3A%22S3XKT4P5%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Xiao%20et%20al.%22%2C%22parsedDate%22%3A%222024-02%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EB.%20Xiao%20%3Ci%3Eet%20al.%3C%5C%2Fi%3E%2C%20%5C%22%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_TVT2024.pdf%27%3EStochastic%20graph%20neural%20network-based%20value%20decomposition%20for%20MARL%20in%20Internet%20of%20vehicles%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EIEEE%20%20Trans.%20Veh.%20Tech.%3C%5C%2Fi%3E%2C%20vol.%2073%2C%20no.%202%2C%20pp.%201582%5Cu20131596%2C%20Feb.%202024%2C%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DS3XKT4P5%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Stochastic%20graph%20neural%20network-based%20value%20decomposition%20for%20MARL%20in%20Internet%20of%20vehicles%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Baidi%22%2C%22lastName%22%3A%22Xiao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Fei%22%2C%22lastName%22%3A%22Wang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Chenghui%22%2C%22lastName%22%3A%22Peng%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jianjun%22%2C%22lastName%22%3A%22Wu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22Autonomous%20driving%20has%20witnessed%20incredible%20advances%20in%20the%20past%20several%20decades%2C%20while%20Multi-Agent%20Reinforcement%20Learning%20%28MARL%29%20promises%20to%20satisfy%20the%20essential%20need%20of%20autonomous%20vehicle%20control%20in%20a%20wireless%20connected%20vehicle%20networks.%20In%20MARL%2C%20how%20to%20effectively%20decompose%20a%20global%20feedback%20into%20the%20relative%20contributions%20of%20individual%20agents%20belongs%20to%20one%20of%20the%20most%20fundamental%20problems.%20However%2C%20the%20environment%20volatility%20due%20to%20vehicle%20movement%20and%20wireless%20disturbance%20could%20significantly%20shape%20time-varying%20topological%20relationships%20among%20agents%2C%20thus%20making%20the%20Value%20Decomposition%20%28VD%29%20challenging.%20Therefore%2C%20in%20order%20to%20cope%20with%20this%20annoying%20volatility%2C%20it%20becomes%20imperative%20to%20design%20a%20dynamic%20VD%20framework.%20Hence%2C%20in%20this%20paper%2C%20we%20propose%20a%20novel%20Stochastic%20Value%20Mixing%20%28SVMIX%29%20methodology%20by%20taking%20account%20of%20dynamic%20topological%20features%20during%20VD%20and%20incorporating%20the%20corresponding%20components%20into%20a%20multi-agent%20actor-critic%20architecture.%20In%20particular%2C%20Stochastic%20Graph%20Neural%20Network%20%28SGNN%29%20is%20leveraged%20to%20effectively%20capture%20underlying%20dynamics%20in%20topological%20features%20and%20improve%20the%20flexibility%20of%20VD%20against%20the%20environment%20volatility.%20Finally%2C%20the%20superiority%20of%20SVMIX%20is%20verified%20through%20extensive%20simulations.%22%2C%22date%22%3A%222024-2%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_TVT2024.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222024-02-27T06%3A51%3A02Z%22%7D%7D%2C%7B%22key%22%3A%22U8DIPWZN%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Chen%20et%20al.%22%2C%22parsedDate%22%3A%222024%22%2C%22numChildren%22%3A0%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EY.%20Chen%20%3Ci%3Eet%20al.%3C%5C%2Fi%3E%2C%20%26%23x201C%3BNetGPT%3A%20An%20AI-native%20network%20architecture%20for%20provisioning%20beyond%20personalized%20generative%20services%2C%26%23x201D%3B%20%3Ci%3EIEEE%20Network%3C%5C%2Fi%3E%2C%202024.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DU8DIPWZN%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22NetGPT%3A%20An%20AI-native%20network%20architecture%20for%20provisioning%20beyond%20personalized%20generative%20services%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yuxuan%22%2C%22lastName%22%3A%22Chen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Chenghui%22%2C%22lastName%22%3A%22Peng%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jianjun%22%2C%22lastName%22%3A%22Wu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Ekram%22%2C%22lastName%22%3A%22Hossain%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%222024%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%5D%2C%22dateModified%22%3A%222024-03-17T08%3A15%3A45Z%22%7D%7D%2C%7B%22key%22%3A%22CXAHHYNF%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Xu%20et%20al.%22%2C%22parsedDate%22%3A%222024%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EX.%20Xu%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%5C%22%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_TWC2024.pdf%27%3EThe%20gradient%20convergence%20bound%20of%20federated%20multi-agent%20reinforcement%20learning%20with%20efficient%20communication%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EIEEE%20Trans.%20Wireless%20Commun.%3C%5C%2Fi%3E%2C%20vol.%2023%2C%20no.%201%2C%20pp.%20507%5Cu2013528%2C%202024%2C%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DCXAHHYNF%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22The%20gradient%20convergence%20bound%20of%20federated%20multi-agent%20reinforcement%20learning%20with%20efficient%20communication%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xing%22%2C%22lastName%22%3A%22Xu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22The%20paper%20considers%20independent%20reinforcement%20learning%20%28IRL%29%20for%20multi-agent%20collaborative%20decision-making%20in%20the%20paradigm%20of%20federated%20learning%20%28FL%29.%20However%2C%20FL%20generates%20excessive%20communication%20overheads%20between%20agents%20and%20a%20remote%20central%20server%2C%20especially%20when%20it%20involves%20a%20large%20number%20of%20agents%20or%20iterations.%20Besides%2C%20due%20to%20the%20heterogeneity%20of%20independent%20learning%20environments%2C%20multiple%20agents%20may%20undergo%20asynchronous%20Markov%20decision%20processes%20%28MDPs%29%2C%20which%20will%20affect%20the%20training%20samples%20and%20the%20model%27s%20convergence%20performance.%20On%20top%20of%20the%20variation-aware%20periodic%20averaging%20%28VPA%29%20method%20and%20the%20policy-based%20deep%20reinforcement%20learning%20%28DRL%29%20algorithm%20%28i.e.%2C%20proximal%20policy%20optimization%20%28PPO%29%29%2C%20this%20paper%20proposes%20two%20advanced%20optimization%20schemes%20orienting%20to%20stochastic%20gradient%20descent%20%28SGD%29%3A%201%29%20A%20decay-based%20scheme%20gradually%20decays%20the%20weights%20of%20a%20model%27s%20local%20gradients%20with%20the%20progress%20of%20successive%20local%20updates%2C%20and%202%29%20By%20representing%20the%20agents%20as%20a%20graph%2C%20a%20consensus-based%20scheme%20studies%20the%20impact%20of%20exchanging%20a%20model%27s%20local%20gradients%20among%20nearby%20agents%20from%20an%20algebraic%20connectivity%20perspective.%20This%20paper%20also%20provides%20novel%20convergence%20guarantees%20for%20both%20developed%20schemes%2C%20and%20demonstrates%20their%20superior%20effectiveness%20and%20efficiency%20in%20improving%20the%20system%27s%20utility%20value%20through%20theoretical%20analyses%20and%20simulation%20results.%22%2C%22date%22%3A%222024%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_TWC2024.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222024-02-27T06%3A51%3A13Z%22%7D%7D%2C%7B%22key%22%3A%22TML62W6B%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Li%20et%20al.%22%2C%22parsedDate%22%3A%222023-12%22%2C%22numChildren%22%3A0%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3ES.%20Li%2C%20Y.%20Xiang%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BImitation%20learning%20based%20alternative%20multi-agent%20proximal%20policy%20optimization%20for%20well-formed%20swarm-oriented%20pursuit%20avoidance%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20ICCC%202023%3C%5C%2Fi%3E%2C%20Chengdu%2C%20China%2C%20Dec.%202023.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DTML62W6B%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Imitation%20learning%20based%20alternative%20multi-agent%20proximal%20policy%20optimization%20for%20well-formed%20swarm-oriented%20pursuit%20avoidance%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Sizhao%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yuming%22%2C%22lastName%22%3A%22Xiang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%22December%202023%22%2C%22proceedingsTitle%22%3A%22Proc.%20ICCC%202023%22%2C%22conferenceName%22%3A%22ICCC%202023%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222023-11-11T01%3A07%3A40Z%22%7D%7D%2C%7B%22key%22%3A%22FD2C33II%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Yu%20et%20al.%22%2C%22parsedDate%22%3A%222023-12%22%2C%22numChildren%22%3A0%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EX.%20Yu%20%3Ci%3Eet%20al.%3C%5C%2Fi%3E%2C%20%26%23x201C%3BCommunication-efficient%20cooperative%20multi-agent%20PPO%20via%20regulated%20segment%20mixture%20in%20internet%20of%20vehicles%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20IEEE%20Globecom%202023%3C%5C%2Fi%3E%2C%20Kuala%20Lumpur%2C%20Malaysia%2C%20Dec.%202023.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DFD2C33II%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Communication-efficient%20cooperative%20multi-agent%20PPO%20via%20regulated%20segment%20mixture%20in%20internet%20of%20vehicles%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xiaoxue%22%2C%22lastName%22%3A%22Yu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Fei%22%2C%22lastName%22%3A%22Wang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Chenghui%22%2C%22lastName%22%3A%22Peng%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Chengchao%22%2C%22lastName%22%3A%22Liang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%222023-12%22%2C%22proceedingsTitle%22%3A%22Proc.%20IEEE%20Globecom%202023%22%2C%22conferenceName%22%3A%22IEEE%20Globecom%202023%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222023-08-04T09%3A33%3A28Z%22%7D%7D%2C%7B%22key%22%3A%22NKI5IVS4%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Xiang%20et%20al.%22%2C%22parsedDate%22%3A%222023-11%22%2C%22numChildren%22%3A0%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EY.%20Xiang%2C%20S.%20Li%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BDecentralized%20adaptive%20formation%20via%20consensus-oriented%20multi-agent%20communication%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20WCSP%202023%3C%5C%2Fi%3E%2C%20Hangzhou%2C%20China%2C%20Nov.%202023.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DNKI5IVS4%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Decentralized%20adaptive%20formation%20via%20consensus-oriented%20multi-agent%20communication%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yuming%22%2C%22lastName%22%3A%22Xiang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Sizhao%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%22November%202023%22%2C%22proceedingsTitle%22%3A%22Proc.%20WCSP%202023%22%2C%22conferenceName%22%3A%22WCSP%202023%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222023-09-12T07%3A53%3A06Z%22%7D%7D%2C%7B%22key%22%3A%222WNCETHJ%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Geng%20et%20al.%22%2C%22parsedDate%22%3A%222023-11%22%2C%22numChildren%22%3A0%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EW.%20Geng%2C%20B.%20Xiao%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20N.%20Wei%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BDecomposition-based%20multi-agent%20distributional%20reinforcement%20learning%20for%20task-oriented%20UAV%20collaboration%20with%20noisy%20rewards%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20WCSP%202023%3C%5C%2Fi%3E%2C%20Hangzhou%2C%20China%2C%20Nov.%202023.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3D2WNCETHJ%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Decomposition-based%20multi-agent%20distributional%20reinforcement%20learning%20for%20task-oriented%20UAV%20collaboration%20with%20noisy%20rewards%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Wei%22%2C%22lastName%22%3A%22Geng%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Baidi%22%2C%22lastName%22%3A%22Xiao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Ning%22%2C%22lastName%22%3A%22Wei%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%22November%202023%22%2C%22proceedingsTitle%22%3A%22Proc.%20WCSP%202023%22%2C%22conferenceName%22%3A%22WCSP%202023%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222023-09-12T07%3A52%3A12Z%22%7D%7D%2C%7B%22key%22%3A%225JX9BFAM%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Ni%20et%20al.%22%2C%22parsedDate%22%3A%222023-09%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EF.%20Ni%2C%20Z.%20Zhao%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20and%20H.%20Zhang%2C%20%5C%22%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fieeexplore.ieee.org%5C%2Fstamp%5C%2Fstamp.jsp%3Ftp%3D%26arnumber%3D10232997%27%3EEEG%20signal-assisted%20algebraic%20topological%20feature-enhanced%20deep%20neural%20networks%20for%20gestalt%20illusory%20contour%20perception%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EIEEE%20Access%3C%5C%2Fi%3E%2C%20vol.%2011%2C%20pp.%2096029%5Cu201396042%2C%20Sep.%202023%2C%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3D5JX9BFAM%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22EEG%20signal-assisted%20algebraic%20topological%20feature-enhanced%20deep%20neural%20networks%20for%20gestalt%20illusory%20contour%20perception%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Fei%22%2C%22lastName%22%3A%22Ni%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22Deep%20neural%20networks%20%28DNNs%29%20have%20been%20incredibly%20successful%20at%20correctly%20classifying%20various%20input%20types%2C%20including%20images%2C%20speech%2C%20and%20data%2C%20according%20to%20consumer%20preferences.%20However%2C%20the%20efficacy%20of%20DNNs%20in%20Gestalt%20classification%20tasks%2C%20an%20important%20case%20in%20manifesting%20human%20perceptual%20capability%2C%20remains%20challenging%2C%20as%20DNNs%20are%20generally%20unable%20to%20perceive%20illusory%20closure%20from%20Gestalt%20images%20unless%20DNNs%20are%20carefully%20calibrated%20using%20a%20significant%20amount%20of%20priori%20information.%20However%2C%20altering%20the%20input%2C%20as%20minute%20changes%20typically%20imperceptible%20to%20humans%2C%20can%20confound%20carefully%20calibrated%20DNNs.%20In%20this%20study%2C%20EEG%20signal-empowered%20deep%20clustering%20based%20on%20topological%20data%20analysis%20%28TDA%29%20is%20proposed.%20The%20adoption%20of%20TDA%20manifests%20the%20separability%20of%20EEG%20signals%20responsive%20to%20Gestalt%20and%20non-Gestalt%20images.%20It%20also%20yields%20new%20families%20of%20features%20and%20descriptors%20for%20Gestalt%20illusory%20contours%20by%20extracting%20topological%20and%20geometric%20information.%20Furthermore%2C%20the%20combination%20of%20analyzed%20EEG%20signals%20and%20digital%20images%20further%20benefits%20the%20recognition%20of%20Gestalt%20illusory%20contours.%20Extensive%20experiments%20have%20shown%20convincing%20improvements%20over%20the%20state-of-the-art%20DNNs%20%28e.g.%2C%20DeepCluster%20and%20multi-view%20clustering%20%5BMVC%5D%29.%20In%20particular%2C%20DeepCluster%20with%20TDA%20can%20perceive%20illusory%20contours%20to%20some%20degree%2C%20given%20its%2066.5%25%20classification%20accuracy.%20Nevertheless%2C%20on%20top%20of%20extracted%20topological%20features%20from%20EEG%20signals%2C%20it%20produces%20higher%20classification%20accuracy%20%28i.e.%2C%2071.9%25%29%2C%20indicating%20the%20features%20extracted%20from%20the%20EEG%20signal%20contribute%20to%20perceive%20Gestalt%20illusory%20contours.%20On%20the%20other%20hand%2C%20the%20applications%20of%20topological%20features%20to%20the%20latest%20MVC%20also%20bring%204.5%25%20improvement%20and%20demonstrate%20the%20effectiveness.%22%2C%22date%22%3A%222023-9%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fieeexplore.ieee.org%5C%2Fstamp%5C%2Fstamp.jsp%3Ftp%3D%26arnumber%3D10232997%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222024-02-27T06%3A53%3A39Z%22%7D%7D%2C%7B%22key%22%3A%22EM7EHJKF%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Tong%20et%20al.%22%2C%22parsedDate%22%3A%222023-09%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3ES.%20Tong%2C%20X.%20Yu%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20K.%20Lu%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BAlternate%20learning%20based%20sparse%20semantic%20communications%20for%20visual%20transmission%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20IEEE%20PIMRC%202023%3C%5C%2Fi%3E%2C%20Toronto%2C%20Canada%2C%20Sep.%202023.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DEM7EHJKF%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Alternate%20learning%20based%20sparse%20semantic%20communications%20for%20visual%20transmission%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Siyu%22%2C%22lastName%22%3A%22Tong%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xiaoxue%22%2C%22lastName%22%3A%22Yu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Kun%22%2C%22lastName%22%3A%22Lu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22Semantic%20communication%20%28SemCom%29%20demonstrates%20strong%20superiority%20over%20conventional%20bit-level%20accurate%20transmission%2C%20by%20only%20attempting%20to%20recover%20the%20essential%20semantic%20information%20of%20data.%20In%20this%20paper%2C%20in%20order%20to%20tackle%20the%20non-differentiability%20of%20channels%2C%20we%20propose%20an%20alternate%20learning%20based%20SemCom%20system%20for%20visual%20transmission%2C%20named%20SparseSBC.%20Specially%2C%20SparseSBC%20leverages%20two%20separate%20Deep%20Neural%20Network%20%28DNN%29-based%20models%20at%20the%20transmitter%20and%20receiver%2C%20respectively%2C%20and%20learns%20the%20encoding%20and%20decoding%20in%20an%20alternate%20manner%2C%20rather%20than%20the%20joint%20optimization%20in%20existing%20literature%2C%20so%20as%20to%20solving%20the%20non-differentiability%20in%20the%20channel.%20In%20particular%2C%20a%20%60%60self-critic%5C%22%20training%20scheme%20is%20leveraged%20for%20stable%20training.%20Moreover%2C%20the%20DNN-based%20transmitter%20generates%20a%20sparse%20set%20of%20bits%20in%20deduced%20%60%60semantic%20bases%5C%22%2C%20by%20further%20incorporating%20a%20binary%20quantization%20module%20on%20the%20basis%20of%20minimal%20detrimental%20effect%20to%20the%20semantic%20accuracy.%20Extensive%20simulation%20results%20validate%20that%20SparseSBC%20shows%20efficient%20and%20effective%20transmission%20performance%20under%20various%20channel%20conditions%2C%20and%20outperforms%20typical%20SemCom%20solutions.%22%2C%22date%22%3A%222023-9%22%2C%22proceedingsTitle%22%3A%22Proc.%20IEEE%20PIMRC%202023%22%2C%22conferenceName%22%3A%22IEEE%20PIMRC%202023%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222023-08-04T09%3A36%3A29Z%22%7D%7D%2C%7B%22key%22%3A%22Q6TKJGPB%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Liu%20et%20al.%22%2C%22parsedDate%22%3A%222023-09%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EY.%20Liu%2C%20D.%20Li%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20Y.%20Zhu%2C%20and%20H.%20Zhang%2C%20%5C%22%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_COMMAG2023.pdf%27%3ESecure%20and%20efficient%20stigmergy-empowered%20blockchain%20framework%20for%20heterogeneous%20collaborative%20services%20in%20the%20Internet%20of%20vehicles%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EIEEE%20Commun.%20Mag.%3C%5C%2Fi%3E%2C%20vol.%2061%2C%20no.%209%2C%20pp.%20186--%20192%2C%20Sep.%202023%2C%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DQ6TKJGPB%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Secure%20and%20efficient%20stigmergy-empowered%20blockchain%20framework%20for%20heterogeneous%20collaborative%20services%20in%20the%20Internet%20of%20vehicles%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yuntao%22%2C%22lastName%22%3A%22Liu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Daoxun%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yongdong%22%2C%22lastName%22%3A%22Zhu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22The%20Internet%20of%20Vehicles%20%28IoV%29%20is%20expected%20to%20ad-%20dress%20the%20significant%20problems%20of%20modern%20transportation%20through%20collaboration%20among%20various%20entities.%20It%20is%20crucial%20to%20establish%20a%20collaborative%20mechanism%20for%20untrusted%20entities%20to%20achieve%20the%20full%20potential%20of%20IoV.%20Blockchain%20is%20a%20promising%20solution%20for%20building%20a%20credible%20environment%20for%20entities.%20However%2C%20because%20of%20the%20difference%20and%20complexity%20of%20services%2C%20using%20a%20classical%20blockchain%20system%20to%20support%20heterogeneous%20collaborative%20ser-%20vices%20in%20IoV%20causes%20some%20challenges%20in%20smart%20contract%20support%2C%20system%20security%2C%20and%20computational%20efficiency.%20In%20this%20article%2C%20we%20propose%20a%20stigmergy-empowered%20blockchain%20framework%20called%20SEB%2C%20which%20enables%20untrusted%20IoV%20entities%20to%20perform%20heteroge-%20neous%20collaborative%20services%20conveniently%2C%20securely%2C%20and%20efficiently.%20Specifically%2C%20we%20first%20explore%20the%20characteristics%20of%20collabora-%20tive%20services%20and%20analyze%20the%20challenges%20of%20existing%20blockchain%20systems.%20Furthermore%2C%20we%20introduce%20the%20stigmergy%20of%20swarm%20intelligence%20into%20blockchain%20and%20integrate%20the%20stigmergy%20into%20SEB%20by%20designing%20a%20new%20transaction%20data%20structure%2C%20a%20digital%20pheromone%20and%20transaction%20selection%20rules%2C%20and%20a%20new%20transaction%20selection%20algorithm.%20Simulation%20experiments%20demonstrate%20that%20compared%20with%20IOTA%2C%20SEB%20can%20reduce%20smart%20contract%20transaction%20searches%20by%20about%2056%25%2C%20increase%20the%20average%20chain%20length%20by%20about%2083%25%2C%20and%20reduce%20the%20computation%20time%20of%20the%20transaction%20selection%20algorithm%20by%20up%20to%2098%25.%22%2C%22date%22%3A%222023-9%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_COMMAG2023.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222024-02-27T06%3A50%3A05Z%22%7D%7D%2C%7B%22key%22%3A%22WPD6FWKH%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Wang%20et%20al.%22%2C%22parsedDate%22%3A%222023-07%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EB.%20Wang%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20J.%20Zhu%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BKnowledge%20enhanced%20semantic%20communication%20receiver%2C%26%23x201D%3B%20%3Ci%3EIEEE%20Commun.%20Lett.%3C%5C%2Fi%3E%2C%20vol.%2027%2C%20no.%207%2C%20pp.%201794%26%23x2013%3B1798%2C%20Jul.%202023.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DWPD6FWKH%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Knowledge%20enhanced%20semantic%20communication%20receiver%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Bingyan%22%2C%22lastName%22%3A%22Wang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jianhang%22%2C%22lastName%22%3A%22Zhu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22In%20recent%20years%2C%20with%20the%20rapid%20development%20of%20deep%20learning%20and%20natural%20language%20processing%20technologies%2C%20semantic%20communication%20has%20become%20a%20topic%20of%20great%20interest%20in%20the%20field%20of%20communication.%20Although%20existing%20deep%20learning-based%20semantic%20communication%20approaches%20have%20shown%20many%20advantages%2C%20they%20still%20do%20not%20make%20sufficient%20use%20of%20prior%20knowledge.%20Moreover%2C%20most%20existing%20semantic%20communication%20methods%20focus%20on%20the%20semantic%20encoding%20at%20the%20transmitter%20side%2C%20while%20we%20believe%20that%20the%20semantic%20decoding%20capability%20of%20the%20receiver%20should%20also%20be%20concerned.%20In%20this%20paper%2C%20we%20propose%20a%20knowledge%20enhanced%20semantic%20communication%20framework%20in%20which%20the%20receiver%20can%20more%20actively%20utilize%20the%20facts%20in%20the%20knowledge%20base%20for%20semantic%20reasoning%20and%20decoding%2C%20on%20the%20basis%20of%20only%20affecting%20the%20parameters%20rather%20than%20the%20structure%20of%20the%20neural%20networks%20at%20the%20transmitter%20side.%20Specifically%2C%20we%20design%20a%20transformer-%20based%20knowledge%20extractor%20to%20find%20relevant%20factual%20triples%20for%20the%20received%20noisy%20signal.%20Extensive%20simulation%20results%20on%20the%20WebNLG%20dataset%20demonstrate%20that%20the%20proposed%20receiver%20yields%20superior%20performance%20on%20top%20of%20the%20knowledge%20graph%20enhanced%20decoding.%22%2C%22date%22%3A%222023-7%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222023-09-16T06%3A26%3A10Z%22%7D%7D%2C%7B%22key%22%3A%22MXM5RNMM%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Zhu%20et%20al.%22%2C%22parsedDate%22%3A%222023-06%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EJ.%20Zhu%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20X.%20Chen%2C%20S.%20Mao%2C%20J.%20Wu%2C%20and%20Z.%20Zhao%2C%20%26%23x201C%3BSemantics-enhanced%20temporal%20graph%20networks%20for%20content%20caching%20and%20energy%20saving%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20IEEE%20ICC%202023%3C%5C%2Fi%3E%2C%20Rome%2C%20Italy%2C%20Jun.%202023.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DMXM5RNMM%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Semantics-enhanced%20temporal%20graph%20networks%20for%20content%20caching%20and%20energy%20saving%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jianhang%22%2C%22lastName%22%3A%22Zhu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xianfu%22%2C%22lastName%22%3A%22Chen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Shiwen%22%2C%22lastName%22%3A%22Mao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jianjun%22%2C%22lastName%22%3A%22Wu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%5D%2C%22abstractNote%22%3A%22Only%20the%20chairs%20can%20edit%20The%20enormous%20amount%20of%20network%20equipment%20and%20users%20implies%20a%20tremendous%20growth%20of%20Internet%20traffic%20for%20multimedia%20services.%20To%20mitigate%20the%20traffic%20pressure%2C%20architectures%20with%20in-network%20storage%20have%20been%20proposed%20to%20cache%20popular%20content%20at%20devices%20in%20close%20proximity%20to%20users%20in%20order%20to%20decrease%20the%20number%20of%20backhaul%20hops.%20Meanwhile%2C%20the%20reduced%20transmission%20distance%20also%20contributes%20to%20energy%20saving.%20However%2C%20due%20to%20limited%20storage%2C%20only%20a%20fraction%20of%20the%20content%20can%20be%20cached%2C%20while%20caching%20the%20most%20popular%20content%20is%20cost-effective.%20Correspondingly%2C%20it%20becomes%20essential%20to%20devise%20an%20effective%20popularity%20prediction%20method.%20In%20this%20regard%2C%20some%20existing%20efforts%20manifest%20the%20effectiveness%20of%20dynamic%20graph%20neural%20network%20%28DGNN%29%20models%2C%20but%20it%20remains%20challenging%20to%20tackle%20sparse%20datasets.%20Herein%2C%20we%20first%20propose%20a%20reformative%20temporal%20graph%20network%2C%20named%20STGN%2C%20to%20address%20the%20challenge%20and%20improve%20prediction%20performance.%20Specifically%2C%20the%20STGN%20model%20leverages%20extra%20semantic%20messages%20to%20help%20establish%20implicit%20paths%20within%20the%20sparse%20interaction%20graph%20and%20enhance%20the%20temporal%20and%20structural%20learning%20of%20a%20DGNN%20model.%20Furthermore%2C%20we%20devise%20a%20user-specific%20attention%20mechanism%20to%20aggregate%20various%20semantics%20in%20a%20fine-grained%20manner.%20Finally%2C%20extensive%20simulations%20verify%20the%20superiority%20of%20our%20STGN%20models%20and%20demonstrate%20the%20potential%20in%20terms%20of%20energy-saving.%22%2C%22date%22%3A%222023-6%22%2C%22proceedingsTitle%22%3A%22Proc.%20IEEE%20ICC%202023%22%2C%22conferenceName%22%3A%22IEEE%20ICC%202023%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222023-04-19T12%3A42%3A36Z%22%7D%7D%2C%7B%22key%22%3A%22LIMHNU44%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Fang%20et%20al.%22%2C%22parsedDate%22%3A%222023-06%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EJ.%20Fang%2C%20C.%20Wang%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20H.%20Wei%2C%20and%20M.%20Zhao%2C%20%26%23x201C%3BMatrix%20factorization%20and%20deep%20autoencoder%20based%20clustering%20scheme%20for%20large-scale%20UAV%20networks%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20IEEE%20VTC%202023-Spring%3C%5C%2Fi%3E%2C%20Florence%2C%20Italy%2C%20Jun.%202023.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DLIMHNU44%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Matrix%20factorization%20and%20deep%20autoencoder%20based%20clustering%20scheme%20for%20large-scale%20UAV%20networks%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jiaolan%22%2C%22lastName%22%3A%22Fang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Chan%22%2C%22lastName%22%3A%22Wang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Hanyu%22%2C%22lastName%22%3A%22Wei%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Minjian%22%2C%22lastName%22%3A%22Zhao%22%7D%5D%2C%22abstractNote%22%3A%22In%20recent%20years%2C%20unmanned%20aerial%20vehicle%20%28UAV%29%20ad-%20hoc%20networks%20have%20achieved%20rapid%20development%20due%20to%20their%20autonomy%20and%20high%20reliability.%20Typically%2C%20clustering%20is%20widely%20adopted%20to%20reduce%20the%20degradation%20of%20network%20performance%20in%20large-scale%20UAV%20networks.%20However%2C%20due%20to%20the%20limited%20channel%20resources%20and%20complex%20communication%20environment%2C%20it%20becomes%20challenging%20to%20obtain%20complete%20information%20needed%20for%20clustering%20from%20remote%20nodes.%20In%20addition%2C%20the%20high%20dimension%20and%20non-%20linear%20relationships%20of%20data%20also%20make%20large-scale%20clustering%20diffcult.%20Therefore%2C%20in%20this%20paper%2C%20we%20propose%20a%20noval%20distributed%20clustering%20scheme.%20First%2C%20a%20matrix%20factorization%20%28MF%29%20based%20time%20series%20algorithm%20is%20proposed%20to%20predict%20and%20complement%20the%20incomplete%20information.%20Second%2C%20we%20adopt%20deep%20autoencoder%20to%20wisely%20incorporate%20the%20non-linear%20relationship%20of%20information%20needed%20for%20clustering.%20Finally%20we%20extract%20features%20as%20input%20to%20k-%20means%20to%20obtain%20clustering%20results.%20Simulation%20results%20demonstrate%20the%20effectiveness%20of%20the%20proposed%20clustering%20scheme.%22%2C%22date%22%3A%222023-6%22%2C%22proceedingsTitle%22%3A%22Proc.%20IEEE%20VTC%202023-Spring%22%2C%22conferenceName%22%3A%22IEEE%20VTC%202023-Spring%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222023-07-03T09%3A35%3A04Z%22%7D%7D%2C%7B%22key%22%3A%228WYQPY9L%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Xiao%20et%20al.%22%2C%22parsedDate%22%3A%222023-06%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EB.%20Xiao%20%3Ci%3Eet%20al.%3C%5C%2Fi%3E%2C%20%26%23x201C%3BStochastic%20graph%20neural%20network-based%20value%20decomposition%20for%20multi-agent%20reinforcement%20learning%20in%20urban%20traffic%20control%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20IEEE%20VTC%202023-Spring%3C%5C%2Fi%3E%2C%20Florence%2C%20Italy%2C%20Jun.%202023.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3D8WYQPY9L%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Stochastic%20graph%20neural%20network-based%20value%20decomposition%20for%20multi-agent%20reinforcement%20learning%20in%20urban%20traffic%20control%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Baidi%22%2C%22lastName%22%3A%22Xiao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Fei%22%2C%22lastName%22%3A%22Wang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Chenghui%22%2C%22lastName%22%3A%22Peng%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jianjun%22%2C%22lastName%22%3A%22Wu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22Multi-Agent%20Reinforcement%20Learning%20%28MARL%29%20has%20reached%20astonishing%20achievements%20in%20various%20fields%20such%20as%20the%20traffic%20control%20of%20vehicles%20in%20a%20wireless%20connected%20environment.%20In%20MARL%2C%20how%20to%20effectively%20decompose%20a%20global%20feedback%20into%20the%20relative%20contributions%20of%20individual%20agents%20belongs%20to%20one%20of%20the%20most%20fundamental%20problems.%20However%2C%20the%20volatility%20of%20the%20environment%20%28%5C%5Ceg%20the%20vehicle%20movement%20and%20wireless%20disturbance%29%20could%20significantly%20shape%20the%20time-varying%20topological%20relationships%20among%20agents%2C%20thus%20making%20the%20Value%20Decomposition%20%28VD%29%20challenging.%20Therefore%2C%20in%20order%20to%20cope%20with%20this%20annoying%20volatility%2C%20it%20becomes%20imperative%20to%20design%20a%20dynamic%20VD%20framework.%20Hence%2C%20in%20this%20paper%2C%20we%20propose%20a%20novel%20Stochastic%20VMIX%20%28SVMIX%29%20methodology%20by%20embedding%20the%20dynamic%20topological%20features%20into%20the%20VD%20and%20incorporating%20the%20corresponding%20components%20into%20a%20multi-agent%20actor-critic%20architecture.%20In%20particular%2C%20the%20Stochastic%20Graph%20Neural%20Network%20%28SGNN%29%20is%20leveraged%20to%20effectively%20extract%20underlying%20dynamics%20embedded%20in%20topological%20features%20and%20improve%20the%20flexibility%20of%20VD%20against%20the%20environment%20volatility.%20Finally%2C%20the%20superiority%20of%20SVMIX%20is%20verified%20through%20extensive%20simulations.%22%2C%22date%22%3A%222023-6%22%2C%22proceedingsTitle%22%3A%22Proc.%20IEEE%20VTC%202023-Spring%22%2C%22conferenceName%22%3A%22IEEE%20VTC%202023-Spring%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222023-04-19T12%3A45%3A04Z%22%7D%7D%2C%7B%22key%22%3A%222BSDLKQV%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Zhu%20et%20al.%22%2C%22parsedDate%22%3A%222023-04%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EJ.%20Zhu%20%3Ci%3Eet%20al.%3C%5C%2Fi%3E%2C%20%5C%22%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_TCCN2023.pdf%27%3EAoI-based%20temporal%20attention%20graph%20neural%20network%20for%20popularity%20prediction%20and%20content%20caching%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EIEEE%20Trans.%20Cog.%20Commun.%20Netw.%3C%5C%2Fi%3E%2C%20vol.%209%2C%20no.%202%2C%20pp.%20345%5Cu2013358%2C%20Apr.%202023%2C%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3D2BSDLKQV%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22AoI-based%20temporal%20attention%20graph%20neural%20network%20for%20popularity%20prediction%20and%20content%20caching%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jianhang%22%2C%22lastName%22%3A%22Zhu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Guoru%22%2C%22lastName%22%3A%22Ding%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Chan%22%2C%22lastName%22%3A%22Wang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jianjun%22%2C%22lastName%22%3A%22Wu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22Along%20with%20the%20fast%20development%20of%20network%20technology%20and%20the%20rapid%20growth%20of%20network%20equipment%2C%20the%20data%20throughput%20is%20sharply%20increasing.%20To%20handle%20the%20problem%20of%20backhaul%20bottleneck%20in%20cellular%20network%20and%20satisfy%20people%5Cu2019s%20requirements%20about%20latency%2C%20the%20network%20architecture%2C%20like%20the%20information-centric%20network%20%28ICN%29%20intends%20to%20proactively%20keep%20limited%20popular%20content%20at%20the%20edge%20of%20network%20based%20on%20predicted%20results.%20Meanwhile%2C%20the%20interactions%20between%20the%20content%20%28e.g.%2C%20deep%20neural%20network%20models%2C%20Wikipedia-alike%20knowledge%20base%29%20and%20users%20could%20be%20regarded%20as%20a%20dynamic%20bipartite%20graph.%20In%20this%20paper%2C%20to%20maximize%20the%20cache%20hit%20rate%2C%20we%20leverage%20an%20effective%20dynamic%20graph%20neural%20network%20%28DGNN%29%20to%20jointly%20learn%20the%20structural%20and%20temporal%20patterns%20embedded%20in%20the%20bipartite%20graph.%20Furthermore%2C%20in%20order%20to%20have%20deeper%20insights%20into%20the%20dynamics%20within%20the%20evolving%20graph%2C%20we%20propose%20an%20age%20of%20information%20%28AoI%29%20based%20attention%20mechanism%20to%20extract%20valuable%20historical%20information%20while%20avoiding%20the%20problem%20of%20message%20staleness.%20Combining%20this%20aforementioned%20prediction%20model%2C%20we%20also%20develop%20a%20cache%20selection%20algorithm%20to%20make%20caching%20decisions%20in%20accordance%20with%20the%20prediction%20results.%20Extensive%20results%20demonstrate%20that%20our%20model%20can%20obtain%20a%20higher%20prediction%20accuracy%20than%20other%20state-of-the-art%20schemes%20in%20two%20real-world%20datasets.%20The%20results%20of%20hit%20rate%20further%20verify%20the%20superiority%20of%20the%20caching%20policy%20based%20on%20our%20proposed%20model%20over%20other%20traditional%20ways.%22%2C%22date%22%3A%222023-4%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_TCCN2023.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222024-02-27T06%3A51%3A35Z%22%7D%7D%2C%7B%22key%22%3A%22VKGNQX4R%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Lu%20et%20al.%22%2C%22parsedDate%22%3A%222023-02%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EK.%20Lu%20%3Ci%3Eet%20al.%3C%5C%2Fi%3E%2C%20%5C%22%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_WCOM2023.pdf%27%3ERethinking%20modern%20communication%20from%20semantic%20coding%20to%20semantic%20communication%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EIEEE%20Wireless%20Commun.%3C%5C%2Fi%3E%2C%20vol.%2030%2C%20no.%201%2C%20pp.%20158%5Cu2013164%2C%20Feb.%202023%2C%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DVKGNQX4R%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Rethinking%20modern%20communication%20from%20semantic%20coding%20to%20semantic%20communication%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Kun%22%2C%22lastName%22%3A%22Lu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Qingyang%22%2C%22lastName%22%3A%22Zhou%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xianfu%22%2C%22lastName%22%3A%22Chen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jianjun%22%2C%22lastName%22%3A%22Wu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22Modern%20communications%20are%20usually%20designed%20to%20pursue%20a%20higher%20bit-level%20precision%20and%20fewer%20bits%20while%20transmitting%20a%20message.%20This%20article%20rethinks%20these%20two%20major%20features%20and%20introduc-%20es%20the%20concept%20and%20advantage%20of%20semantics%20that%20characterizes%20a%20new%20kind%20of%20semantics-aware%20communication%20framework%2C%20incorporating%20both%20the%20semantic%20encoding%20and%20the%20semantic%20commu-%20nication%20problem.%20After%20analyzing%20the%20underlying%20defects%20of%20existing%20semantics-aware%20techniques%2C%20we%20establish%20a%20confidence-based%20distillation%20mechanism%20for%20the%20joint%20semantics-noise%20coding%20%28JSNC%29%20prob-%20lem%20and%20a%20reinforcement%20learning%20%28RL%29%20powered%20semantic%20communication%20paradigm%20that%20endows%20a%20system%20the%20ability%20to%20convey%20the%20semantics%20instead%20of%20pursuing%20the%20bit%20level%20accuracy.%20On%20top%20of%20these%20technical%20contributions%2C%20this%20work%20provides%20a%20new%20insight%20to%20understand%20how%20the%20semantics%20are%20pro-%20cessed%20and%20represented%20in%20a%20semantics-aware%20cod-%20ing%20and%20communication%20system%2C%20and%20verifies%20the%20significant%20benefits%20of%20doing%20so.%20Targeted%20on%20the%20next%20generation%5Cu2019s%20semantics-aware%20communication%2C%20some%20critical%20concerns%20and%20open%20challenges%2C%20such%20as%20the%20information%20overhead%2C%20semantic%20security%2C%20and%20implementation%20cost%20are%20also%20discussed%20and%20envisioned.%22%2C%22date%22%3A%222023-2%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_WCOM2023.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222024-02-27T06%3A54%3A53Z%22%7D%7D%2C%7B%22key%22%3A%22UZ88M2HW%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Zhu%20et%20al.%22%2C%22parsedDate%22%3A%222023%22%2C%22numChildren%22%3A0%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EJ.%20Zhu%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20X.%20Chen%2C%20S.%20Mao%2C%20J.%20Wu%2C%20and%20Z.%20Zhao%2C%20%26%23x201C%3BSemantics-enhanced%20temporal%20graph%20networks%20for%20content%20popularity%20prediction%2C%26%23x201D%3B%20%3Ci%3EIEEE%20Trans.%20Mob.%20Comput.%3C%5C%2Fi%3E%2C%202023%2C%20doi%3A%2010.1109%5C%2FTMC.2023.3349315.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DUZ88M2HW%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Semantics-enhanced%20temporal%20graph%20networks%20for%20content%20popularity%20prediction%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jianhang%22%2C%22lastName%22%3A%22Zhu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xianfu%22%2C%22lastName%22%3A%22Chen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Shiwen%22%2C%22lastName%22%3A%22Mao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jianjun%22%2C%22lastName%22%3A%22Wu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%222023%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%2210.1109%5C%2FTMC.2023.3349315%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222024-01-09T01%3A54%3A06Z%22%7D%7D%2C%7B%22key%22%3A%22E7FX5DCW%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Yang%20et%20al.%22%2C%22parsedDate%22%3A%222023%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EY.%20Yang%20%3Ci%3Eet%20al.%3C%5C%2Fi%3E%2C%20%26%23x201C%3BTask-oriented%206G%20native-AI%20network%20architecture%2C%26%23x201D%3B%20%3Ci%3EIEEE%20Network%3C%5C%2Fi%3E%2C%20pp.%201%26%23x2013%3B1%2C%202023%2C%20doi%3A%2010.1109%5C%2FMNET.2023.3321464.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DE7FX5DCW%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Task-oriented%206G%20native-AI%20network%20architecture%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yang%22%2C%22lastName%22%3A%22Yang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jianjun%22%2C%22lastName%22%3A%22Wu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Tianjiao%22%2C%22lastName%22%3A%22Chen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Chenghui%22%2C%22lastName%22%3A%22Peng%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jun%22%2C%22lastName%22%3A%22Wang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Juan%22%2C%22lastName%22%3A%22Deng%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xiaofeng%22%2C%22lastName%22%3A%22Tao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Guangyi%22%2C%22lastName%22%3A%22Liu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Wenjing%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Li%22%2C%22lastName%22%3A%22Yang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yufeng%22%2C%22lastName%22%3A%22He%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Tingting%22%2C%22lastName%22%3A%22Yang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22A.%20Hamid%22%2C%22lastName%22%3A%22Aghvami%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Frank%22%2C%22lastName%22%3A%22Eliassen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Schahram%22%2C%22lastName%22%3A%22Dustdar%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Dusit%22%2C%22lastName%22%3A%22Niyato%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Wanfei%22%2C%22lastName%22%3A%22Sun%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yang%22%2C%22lastName%22%3A%22Xu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yannan%22%2C%22lastName%22%3A%22Yuan%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jiang%22%2C%22lastName%22%3A%22Xie%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Cuiqin%22%2C%22lastName%22%3A%22Dai%22%7D%5D%2C%22abstractNote%22%3A%22The%20vision%20for%206G%20networks%20is%20to%20offer%20pervasive%20intelligence%20and%20internet%20of%20intelligence%2C%20in%20which%20the%20networks%20natively%20support%20artificial%20intelligence%20%28AI%29%2C%20empower%20smart%20applications%20and%20scenarios%20in%20various%20fields%2C%20and%20create%20a%20%5C%22ubiquitous-intelligence%5C%22%20world.%20In%20this%20vision%2C%20the%20traditional%20session-oriented%20architecture%20cannot%20achieve%20flexible%20per-user%20customization%2C%20ultimate%20performance%2C%20security%20and%20reliability%20required%20by%20future%20AI%20services.%20In%20addition%2C%20users%5Cu2019%20requirements%20for%20personalized%20AI%20services%20may%20become%20a%20key%20feature%20in%20the%20near%20future.%20However%2C%20the%20traditional%20AI%20deployment%20based%20on%20cloud%5C%2Fmobile%20edge%20computing%20%28MEC%29%20has%20limitations%20such%20as%20low%20throughput%2C%20long%20delay%2C%20poor%20privacy%2C%20and%20high%20carbon%20emissions%2C%20resulting%20in%20the%20inability%20to%20provide%20personalized%20quality%20of%20experience%20%28QoE%29%20assurance.%20By%20integrating%20AI%20in%20the%20network%2C%20the%20network%20AI%20has%20more%20advantages%20than%20cloud%5C%2FMEC%20AI%2C%20such%20as%20better%20QoS%20assurance%2C%20lower%20latency%2C%20less%20transmission%20and%20computing%20overhead%2C%20and%20stronger%20security%20and%20privacy.%20Therefore%2C%20this%20article%20proposes%20the%20task-oriented%20native-AI%20network%20architecture%20%28TONA%29%2C%20to%20natively%20support%20the%20network%20AI.%20By%20introducing%20task%20control%20and%20quality%20of%20AI%20services%20%28QoAIS%29%20assurance%20mechanisms%20at%20the%20control%20layer%20of%206G%2C%20the%20TONA%20can%20achieve%20the%20finest%20service%20granularity%20at%20the%20task%20level%20for%20guaranteeing%20every%20user%5Cu2019s%20personalized%20QoE.%22%2C%22date%22%3A%222023%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%2210.1109%5C%2FMNET.2023.3321464%22%2C%22ISSN%22%3A%221558-156X%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%5D%2C%22dateModified%22%3A%222024-02-28T06%3A08%3A59Z%22%7D%7D%2C%7B%22key%22%3A%22Y2RLVQFF%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Wei%20et%20al.%22%2C%22parsedDate%22%3A%222022-12%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EH.%20Wei%2C%20C.%20Wang%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20and%20M.%20Zhao%2C%20%26%23x201C%3BMean-field%20MARL-based%20priority-aware%20CSMA%5C%2FCA%20strategy%20in%20large-scale%20manets%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20IEEE%20Globecom%202022%3C%5C%2Fi%3E%2C%20Rio%20de%20Janeiro%2C%20Brazil%2C%20Dec.%202022.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DY2RLVQFF%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Mean-field%20MARL-based%20priority-aware%20CSMA%5C%2FCA%20strategy%20in%20large-scale%20manets%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Hanyu%22%2C%22lastName%22%3A%22Wei%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Chan%22%2C%22lastName%22%3A%22Wang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Minjian%22%2C%22lastName%22%3A%22Zhao%22%7D%5D%2C%22abstractNote%22%3A%22Mobile%20ad-hoc%20network%20%28MANET%29%20has%20attracted%20extensive%20attention%20in%20both%20civilian%20and%20military%20applications%20with%20nodes%20operating%20over%20bandwidth-constrained%20wireless%20links%20in%20a%20self-organized%20manner.%20Typically%2C%20to%20avoid%20potential%20collisions%2C%20CSMA%5C%2FCA%20method%20is%20widely%20adopted%20in%20MANETs%20and%20deeply%20affects%20the%20provisioning%20quality%20of%20service%20%28QoS%29.%20However%2C%20due%20to%20the%20potentially%20severe%20collisions%20in%20a%20large-scale%20MANET%2C%20current%20CSMA%5C%2FCA%20methods%20fail%20to%20support%20QoS%20effectively%2C%20especially%20for%20scenarios%20with%20different%20priority%20services.%20Here%2C%20inspired%20by%20the%20mean-field%20theory%2C%20we%20propose%20a%20priority-aware%20CSMA%5C%2FCA%20strategy%20to%20use%20the%20higher-priority%20packets%20to%20collect%20the%20queueing%20information%20from%20adjacent%20nodes%20in%20the%20MANET%20in%20a%20piggyback%20manner%20and%20model%20the%20channel%20access%20process%20by%20a%20partial%20observation%20Markov%20decision%20process%20%28POMDP%29.%20Furthermore%2C%20multi-agent%20reinforcement%20learning%20%28MARL%29%20is%20adopted%20to%20allow%20each%20individual%20node%20in%20the%20MANET%20to%20dynamically%20select%20the%20appropriate%20contention%20window%20%28CW%29%20and%20gradually%20compute%20the%20global%20optimal%20policy.%20Finally%2C%20extensive%20simulation%20results%20verify%20significantly%20lower%20delay%20and%20packet%20loss%20rate%20for%20the%20higher-priority%20services%20and%20comparable%20provisioning%20quality%20for%20lower-priority%20services%2C%20which%20reflects%20the%20superiority%20over%20baselines%22%2C%22date%22%3A%222022-12%22%2C%22proceedingsTitle%22%3A%22Proc.%20IEEE%20Globecom%202022%22%2C%22conferenceName%22%3A%22IEEE%20Globecom%202022%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222022-12-09T03%3A41%3A05Z%22%7D%7D%2C%7B%22key%22%3A%226XL8KXZY%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Lu%20et%20al.%22%2C%22parsedDate%22%3A%222022-10%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EK.%20Lu%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BContrastive%20monotonic%20pixel-level%20modulation%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20ECCV%3C%5C%2Fi%3E%2C%20in%20ECCV%20%26%23x2019%3B22.%20Tel-Aviv%2C%20Oct.%202022.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3D6XL8KXZY%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Contrastive%20monotonic%20pixel-level%20modulation%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Kun%22%2C%22lastName%22%3A%22Lu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%22Oct.%202022%22%2C%22proceedingsTitle%22%3A%22Proc.%20ECCV%22%2C%22conferenceName%22%3A%22ECCV%202022%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222023-09-16T06%3A26%3A27Z%22%7D%7D%2C%7B%22key%22%3A%2282329CBH%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Xu%20et%20al.%22%2C%22parsedDate%22%3A%222022-09%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EX.%20Xu%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%5C%22%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_TNNLS2022Stigmergy.pdf%27%3EStigmergic%20independent%20reinforcement%20learning%20for%20multiagent%20collaboration%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EIEEE%20Trans.%20Neural%20Netw.%20Learn.%20Syst.%3C%5C%2Fi%3E%2C%20vol.%2033%2C%20no.%209%2C%20pp.%204285%5Cu20134299%2C%20Sep.%202022%2C%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3D82329CBH%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Stigmergic%20independent%20reinforcement%20learning%20for%20multiagent%20collaboration%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xing%22%2C%22lastName%22%3A%22Xu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22With%20the%20rapid%20evolution%20of%20wireless%20mobile%20devices%2C%20there%20emerges%20an%20increased%20need%20to%20design%20effective%20collaboration%20mechanisms%20between%20intelligent%20agents%20to%20gradually%20approach%20the%20final%20collective%20objective%20by%20continuously%20learning%20from%20the%20environment%20based%20on%20their%20individual%20observations.%20In%20this%20regard%2C%20independent%20reinforcement%20learning%20%28IRL%29%20is%20often%20deployed%20in%20multiagent%20collaboration%20to%20alleviate%20the%20problem%20of%20a%20nonstationary%20learning%20environment.%20However%2C%20behavioral%20strategies%20of%20intelligent%20agents%20in%20IRL%20can%20be%20formulated%20only%20upon%20their%20local%20individual%20observations%20of%20the%20global%20environment%2C%20and%20appropriate%20communication%20mechanisms%20must%20be%20introduced%20to%20reduce%20their%20behavioral%20localities.%20In%20this%20article%2C%20we%20address%20the%20problem%20of%20communication%20between%20intelligent%20agents%20in%20IRL%20by%20jointly%20adopting%20mechanisms%20with%20two%20different%20scales.%20For%20the%20large%20scale%2C%20we%20introduce%20the%20stigmergy%20mechanism%20as%20an%20indirect%20communication%20bridge%20between%20independent%20learning%20agents%2C%20and%20carefully%20design%20a%20mathematical%20method%20to%20indicate%20the%20impact%20of%20digital%20pheromone.%20For%20the%20small%20scale%2C%20we%20propose%20a%20conflict-avoidance%20mechanism%20between%20adjacent%20agents%20by%20implementing%20an%20additionally%20embedded%20neural%20network%20to%20provide%20more%20opportunities%20for%20participants%20with%20higher%20action%20priorities.%20In%20addition%2C%20we%20present%20a%20federal%20training%20method%20to%20effectively%20optimize%20the%20neural%20network%20of%20each%20agent%20in%20a%20decentralized%20manner.%20Finally%2C%20we%20establish%20a%20simulation%20scenario%20in%20which%20a%20number%20of%20mobile%20agents%20in%20a%20certain%20area%20move%20automatically%20to%20form%20a%20specified%20target%20shape.%20Extensive%20simulations%20demonstrate%20the%20effectiveness%20of%20our%20proposed%20method.%22%2C%22date%22%3A%222022-9%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_TNNLS2022Stigmergy.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22IILGBZSM%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222024-03-09T07%3A24%3A33Z%22%7D%7D%2C%7B%22key%22%3A%226D92BW4S%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Yi%20et%20al.%22%2C%22parsedDate%22%3A%222022-09%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EX.%20Yi%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20C.%20Peng%2C%20J.%20Wu%2C%20and%20Z.%20Zhao%2C%20%26%23x201C%3BHFedMTL%3A%20Hierarchical%20federated%20multi-task%20learning%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20IEEE%20PIMRC%202022%3C%5C%2Fi%3E%2C%20Virtual%20Edition%2C%20Sep.%202022.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3D6D92BW4S%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22HFedMTL%3A%20Hierarchical%20federated%20multi-task%20learning%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xingfu%22%2C%22lastName%22%3A%22Yi%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Chenghui%22%2C%22lastName%22%3A%22Peng%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jianjun%22%2C%22lastName%22%3A%22Wu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%222022-9%22%2C%22proceedingsTitle%22%3A%22Proc.%20IEEE%20PIMRC%202022%22%2C%22conferenceName%22%3A%22IEEE%20PIMRC%202022%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222023-09-16T06%3A26%3A20Z%22%7D%7D%2C%7B%22key%22%3A%227TTT3YUX%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Zhou%20et%20al.%22%2C%22parsedDate%22%3A%222022-07%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EQ.%20Zhou%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20Y.%20Xiao%2C%20and%20H.%20Zhang%2C%20%5C%22%3Ca%20href%3D%27http%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2022OJCOMMS_AdaptiveSemanticCom.pdf%27%3EAdaptive%20bit%20rate%20control%20in%20semantic%20communication%20with%20incremental%20knowledge-based%20HARQ%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EIEEE%20Open%20J.%20Commun.%20Soc.%3C%5C%2Fi%3E%2C%20vol.%203%2C%20pp.%201076%5Cu20131089%2C%20Jul.%202022%2C%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3D7TTT3YUX%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Adaptive%20bit%20rate%20control%20in%20semantic%20communication%20with%20incremental%20knowledge-based%20HARQ%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Qingyang%22%2C%22lastName%22%3A%22Zhou%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yong%22%2C%22lastName%22%3A%22Xiao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22Semantic%20communication%20has%20witnessed%20a%20great%20progress%20with%20the%20development%20of%20natural%20language%20processing%20%28NLP%29%20and%20deep%20learning%20%28DL%29.%20Although%20existing%20semantic%20communication%20technologies%20can%20effectively%20reduce%20errors%20in%20semantic%20interpretation%2C%20most%20of%20these%20solutions%20adopt%20a%20fixed%20bit%20length%20structure%2C%20along%20with%20a%20rigid%20transmission%20scheme%2C%20which%20is%20inefficient%20and%20lacks%20scalability%20faced%20with%20different%20meanings%20and%20signal-to-noise%20ratio%20%28SNR%29%20conditions.%20In%20this%20paper%2C%20we%20explore%20the%20impact%20of%20adaptive%20bit%20lengths%20on%20semantic%20coding%20%28SC%29%20under%20various%20channel%20conditions.%20First%2C%20we%20propose%20progressive%20semantic%20hybrid%20automatic%20repeat%20request%20%28HARQ%29%20schemes%20that%20utilize%20incremental%20knowledge%20%28IK%29%20to%20simultaneously%20reduce%20the%20communication%20cost%20and%20semantic%20error.%20On%20top%20of%20this%2C%20we%20design%20a%20novel%20semantic%20encoding%20solution%20with%20multi-bit%20length%20selection.%20In%20this%20fashion%2C%20the%20transmitter%20employs%20a%20policy%20network%20to%20decide%20the%20appropriate%20coding%20rate%2C%20so%20as%20to%20secure%20the%20correct%20information%20delivery%20at%20the%20cost%20of%20minimal%20bits.%20Moreover%2C%20a%20specific%20denoiser%20is%20further%20introduced%20to%20reduce%20the%20semantic%20errors%20encountered%20in%20the%20transmission%20process%20according%20to%20the%20semantic%20characteristics%20of%20context.%20Extensive%20simulation%20results%20have%20been%20conducted%20to%20verify%20the%20effectiveness%20of%20the%20proposed%20solution.%22%2C%22date%22%3A%222022-7%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22http%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2022OJCOMMS_AdaptiveSemanticCom.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222023-09-16T06%3A26%3A19Z%22%7D%7D%2C%7B%22key%22%3A%2297MDIPJI%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Xu%20et%20al.%22%2C%22parsedDate%22%3A%222022-05%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EX.%20Xu%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BCommunication-efficient%20consensus%20mechanism%20for%20federated%20reinforcement%20learning%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20IEEE%20ICC%202022%3C%5C%2Fi%3E%2C%20Seoul%2C%20South%20Korea%2C%20May%202022.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3D97MDIPJI%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Communication-efficient%20consensus%20mechanism%20for%20federated%20reinforcement%20learning%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xing%22%2C%22lastName%22%3A%22Xu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%22May%202022%22%2C%22proceedingsTitle%22%3A%22Proc.%20IEEE%20ICC%202022%22%2C%22conferenceName%22%3A%22IEEE%20ICC%202022%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222023-09-16T06%3A26%3A22Z%22%7D%7D%2C%7B%22key%22%3A%22LTFM7EHE%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Xiao%20et%20al.%22%2C%22parsedDate%22%3A%222022-05%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EB.%20Xiao%2C%20Y.%20Shao%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%5C%22%3Ca%20href%3D%27http%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2022InfocomDemo_NetworkSlicing.pdf%27%3EDEMO%3A%20Deep%20reinforcement%20learning%20for%20resource%20management%20in%20cellular%20network%20slicing%3C%5C%2Fa%3E%2C%5C%22%20in%20%3Ci%3EProc.%20IEEE%20INFOCOM%202022%3C%5C%2Fi%3E%2C%20in%20INFOCOM%20%2722.%20Virtual%20Edition%2C%20May%202022.%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DLTFM7EHE%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22DEMO%3A%20Deep%20reinforcement%20learning%20for%20resource%20management%20in%20cellular%20network%20slicing%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Baidi%22%2C%22lastName%22%3A%22Xiao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yan%22%2C%22lastName%22%3A%22Shao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%222022-5%22%2C%22proceedingsTitle%22%3A%22Proc.%20IEEE%20INFOCOM%202022%22%2C%22conferenceName%22%3A%22IEEE%20INFOCOM%202022%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22http%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2022InfocomDemo_NetworkSlicing.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222022-11-22T01%3A21%3A43Z%22%7D%7D%2C%7B%22key%22%3A%22C7M62MLA%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Xu%20et%20al.%22%2C%22parsedDate%22%3A%222022-04%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EX.%20Xu%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%5C%22%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2022CommLett_TrustableStigmergy.pdf%27%3ETrustable%20policy%20collaboration%20scheme%20for%20multi-agent%20stigmergic%20reinforcement%20learning%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EIEEE%20Commun.%20Lett.%3C%5C%2Fi%3E%2C%20vol.%2026%2C%20no.%204%2C%20pp.%20823%5Cu2013827%2C%20Apr.%202022%2C%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DC7M62MLA%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Trustable%20policy%20collaboration%20scheme%20for%20multi-agent%20stigmergic%20reinforcement%20learning%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xing%22%2C%22lastName%22%3A%22Xu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22In%20this%20letter%2C%20we%20propose%20a%20trustable%20policy%20collaboration%5Cnscheme%20in%20the%20paradigm%20of%20multi-agent%20independent%20reinforcement%5Cnlearning%20%28MAIRL%29.%20This%20trustable%20policy%20collaboration%5Cnscheme%20is%20realized%20by%20directly%20mixing%20the%20policy%20parameters%20of%5Cnhomogeneous%20agents%2C%20for%20which%20an%20upper%20bound%20of%20the%20mixture%5Cnmetric%20is%20derived%20to%20guarantee%20the%20policy%20improvement.%20This%5Cntrustable%20policy%20collaboration%20scheme%20can%20update%20the%20behavioral%5Cnpolicies%20of%20agents%20distributedly%20and%20further%20improve%20the%5Cnperformance%20of%20MAIRL.%20In%20addition%2C%20we%20develop%20a%20practical%5Cnimplementation%20of%20this%20trustable%20policy%20collaboration%20scheme%2C%5Cnand%20verify%20its%20effectiveness%20in%20a%20mixed-autonomy%20traffic%20control%5Cnsimulation%20scenario%20through%20the%20performance%20comparison%20with%5Cnother%20typical%20methods.%22%2C%22date%22%3A%222022-4%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2022CommLett_TrustableStigmergy.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222023-09-16T06%3A26%3A21Z%22%7D%7D%2C%7B%22key%22%3A%22R28T3PP4%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Zhu%20et%20al.%22%2C%22parsedDate%22%3A%222022-04%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EJ.%20Zhu%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BAoI-based%20temporal%20attention%20graph%20neural%20network%20for%20popularity%20prediction%20in%20ICN%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20IEEE%20WCNC%202022%3C%5C%2Fi%3E%2C%20in%20WCNC%20%26%23x2019%3B22.%20Austin%2C%20TX%2C%20USA%2C%20Apr.%202022.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DR28T3PP4%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22AoI-based%20temporal%20attention%20graph%20neural%20network%20for%20popularity%20prediction%20in%20ICN%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jianhang%22%2C%22lastName%22%3A%22Zhu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%222022-4%22%2C%22proceedingsTitle%22%3A%22Proc.%20IEEE%20WCNC%202022%22%2C%22conferenceName%22%3A%22IEEE%20WCNC%202022%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222022-11-22T01%3A21%3A18Z%22%7D%7D%2C%7B%22key%22%3A%22YD34KLAJ%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Zhou%20et%20al.%22%2C%22parsedDate%22%3A%222022-03%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EQ.%20Zhou%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20C.%20Peng%2C%20and%20H.%20Zhang%2C%20%5C%22%3Ca%20href%3D%27http%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2022WComLett_UTTransformer.pdf%27%3ESemantic%20communication%20with%20adaptive%20universal%20transformer%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EIEEE%20Wirel.%20Commun.%20Lett.%3C%5C%2Fi%3E%2C%20vol.%2011%2C%20no.%203%2C%20pp.%20453%5Cu2013457%2C%20Mar.%202022%2C%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DYD34KLAJ%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Semantic%20communication%20with%20adaptive%20universal%20transformer%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Qingyang%22%2C%22lastName%22%3A%22Zhou%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Chenghui%22%2C%22lastName%22%3A%22Peng%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22With%20the%20development%20of%20deep%20learning%20%28DL%29%2C%20natural%20language%20processing%20%28NLP%29%20makes%20it%20possible%20for%20us%20to%20analyze%20and%20understand%20a%20large%20amount%20of%20language%20texts.%20Accordingly%2C%20we%20can%20achieve%20semantic%20communication%20in%20terms%20of%20joint%20semantic%20source%20and%20channel%20coding%20over%20a%20noisy%20channel%20with%20the%20help%20of%20NLP.%20However%2C%20the%20existing%20method%20to%20realize%20this%20goal%20is%20to%20use%20a%20fixed%20transformer%20of%20NLP%20while%20ignoring%20the%20difference%20of%20semantic%20information%20contained%20in%20each%20sentence.%20To%20solve%20this%20problem%2C%20we%20propose%20a%20new%20semantic%20communication%20system%20based%20on%20Universal%20Transformer.%20Compared%20with%20the%20traditional%20transformer%2C%20an%20adaptive%20circulation%20mechanism%20is%20introduced%20in%20the%20Universal%20Transformer.%20Through%20the%20introduction%20of%20the%20circulation%20mechanism%2C%20the%20new%20semantic%20communication%20system%20can%20be%20more%20flexible%20to%20transmit%20sentences%20with%20different%20semantic%20information%2C%20and%20achieve%20better%20end-to-end%20performance%20under%20various%20channel%20conditions.%22%2C%22date%22%3A%222022-3%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22http%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2022WComLett_UTTransformer.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222024-03-09T07%3A24%3A07Z%22%7D%7D%2C%7B%22key%22%3A%22UDWQRC6B%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Chen%20et%20al.%22%2C%22parsedDate%22%3A%222022-01%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EM.%20Chen%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20J.%20Crowcroft%2C%20J.%20Wu%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%5C%22%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2022TCOMTCP_DRL_RR.pdf%27%3ERAN%20information-assisted%20TCP%20congestion%20control%20using%20deep%20reinforcement%20learning%20with%20reward%20redistribution%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EIEEE%20Trans.%20Commun.%3C%5C%2Fi%3E%2C%20vol.%2070%2C%20no.%201%2C%20pp.%20215%5Cu2013230%2C%20Jan.%202022%2C%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DUDWQRC6B%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22RAN%20information-assisted%20TCP%20congestion%20control%20using%20deep%20reinforcement%20learning%20with%20reward%20redistribution%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Minghao%22%2C%22lastName%22%3A%22Chen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jon%22%2C%22lastName%22%3A%22Crowcroft%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jianjun%22%2C%22lastName%22%3A%22Wu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22In%20this%20paper%2C%20we%20aim%20to%20propose%20a%20novel%20transmission%20control%20protocol%20%28TCP%29%20congestion%20control%20method%20from%20a%20cross-layer-based%20perspective%20and%20present%20a%20deep%20reinforcement%20learning%20%28DRL%29-driven%20method%20called%20DRL-3R%20%28DRL%20for%20congestion%20control%20with%20Radio%20access%20network%20information%20and%20Reward%20Redistribution%29%20so%20as%20to%20learn%20the%20TCP%20congestion%20control%20policy%20in%20a%20superior%20manner.%20In%20particular%2C%20we%20incorporate%20the%20RAN%20information%20to%20timely%20grasp%20the%20dynamics%20of%20RAN%2C%20and%20empower%20DRL%20to%20learn%20from%20the%20delayed%20RAN%20information%20feedback%20potentially%20induced%20by%20several%20consecutive%20actions.%20Meanwhile%2C%20we%20relax%20the%20implicit%20assumption%20%28that%20the%20feedback%20to%20one%20specific%20action%20returns%20at%20a%20round-trip-time%20%28RTT%29%20after%20the%20action%20is%20applied%29%20in%20previous%20researches%2C%20by%20redistributing%20the%20rewards%20and%20evaluating%20the%20merits%20of%20actions%20more%20accurately.%20Experiment%20results%20show%20that%20besides%20maintaining%20a%20reasonable%20fairness%2C%20DRL-3R%20significantly%20outperforms%20classical%20congestion%20control%20methods%20%28e.g.%2C%20TCP%20Reno%2C%20Westwood%2C%20Cubic%2C%20BBR%20and%20DRL-CC%29%20on%20network%20utility%20by%20achieving%20a%20higher%20throughput%20while%20reducing%20delay%20in%20various%20network%20environments.%22%2C%22date%22%3A%222022-1%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2022TCOMTCP_DRL_RR.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22B655FXSU%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222023-09-16T06%3A26%3A32Z%22%7D%7D%2C%7B%22key%22%3A%22T2AK79UU%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Qi%20et%20al.%22%2C%22parsedDate%22%3A%222021-12%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EC.%20Qi%20%3Ci%3Eet%20al.%3C%5C%2Fi%3E%2C%20%5C%22%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fasp-eurasipjournals.springeropen.com%5C%2Farticles%5C%2F10.1186%5C%2Fs13634-021-00744-4%27%3EAn%20efficient%20pruning%20scheme%20of%20deep%20neural%20networks%20for%20Internet%20of%20things%20applications%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EEURASIP%20J.%20Adv.%20Signal%20Process.%3C%5C%2Fi%3E%2C%20vol.%202021%2C%20no.%201%2C%20pp.%201%5Cu201321%2C%20Dec.%202021%2C%20Accessed%3A%20Nov.%2010%2C%202021.%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DT2AK79UU%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22An%20efficient%20pruning%20scheme%20of%20deep%20neural%20networks%20for%20Internet%20of%20things%20applications%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Chen%22%2C%22lastName%22%3A%22Qi%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Shibo%22%2C%22lastName%22%3A%22Shen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Qing%22%2C%22lastName%22%3A%22Liu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jing%22%2C%22lastName%22%3A%22Liang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22Nowadays%2C%20deep%20neural%20networks%20%28DNNs%29%20have%20been%20rapidly%20deployed%20to%20realize%20a%20number%20of%20functionalities%20like%20sensing%2C%20imaging%2C%20classification%2C%20recognition%2C%20etc.%20However%2C%20the%20computational-intensive%20requirement%20of%20DNNs%20makes%20it%20difficult%20to%20be%20applicable%20for%20resource-limited%20Internet%20of%20Things%20%28IoT%29%20devices.%20In%20this%20paper%2C%20we%20propose%20a%20novel%20pruning-based%20paradigm%20that%20aims%20to%20reduce%20the%20computational%20cost%20of%20DNNs%2C%20by%20uncovering%20a%20more%20compact%20structure%20and%20learning%20the%20effective%20weights%20therein%2C%20on%20the%20basis%20of%20not%20compromising%20the%20expressive%20capability%20of%20DNNs.%20In%20particular%2C%20our%20algorithm%20can%20achieve%20efficient%20end-to-end%20training%20that%20transfers%20a%20redundant%20neural%20network%20to%20a%20compact%20one%20with%20a%20specifically%20targeted%20compression%20rate%20directly.%20We%20comprehensively%20evaluate%20our%20approach%20on%20various%20representative%20benchmark%20datasets%20and%20compared%20with%20typical%20advanced%20convolutional%20neural%20network%20%28CNN%29%20architectures.%20The%20experimental%20results%20verify%20the%20superior%20performance%20and%20robust%20effectiveness%20of%20our%20scheme.%20For%20example%2C%20when%20pruning%20VGG%20on%20CIFAR-10%2C%20our%20proposed%20scheme%20is%20able%20to%20significantly%20reduce%20its%20FLOPs%20%28floating-point%20operations%29%20and%20number%20of%20parameters%20with%20a%20proportion%20of%2076.2%25%20and%2094.1%25%2C%20respectively%2C%20while%20still%20maintaining%20a%20satisfactory%20accuracy.%20To%20sum%20up%2C%20our%20scheme%20could%20facilitate%20the%20integration%20of%20DNNs%20into%20the%20common%20machine-learning-based%20IoT%20framework%20and%20establish%20distributed%20training%20of%20neural%20networks%20in%20both%20cloud%20and%20edge.%22%2C%22date%22%3A%222021%5C%2F12%22%2C%22language%22%3A%22en%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%221687-6180%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fasp-eurasipjournals.springeropen.com%5C%2Farticles%5C%2F10.1186%5C%2Fs13634-021-00744-4%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222023-09-16T06%3A26%3A41Z%22%7D%7D%2C%7B%22key%22%3A%22EWUSPY69%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Li%20et%20al.%22%2C%22parsedDate%22%3A%222021-12%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3E%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20W.%20Liang%2C%20C.%20Peng%2C%20X.%20An%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BNetwork%20AI%20management%20%26amp%3B%20orchestration%3A%20A%20federated%20multi-task%20learning%20case%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20IEEE%20Globecom%202021%20%28EL5GMNB%20Workshop%29%3C%5C%2Fi%3E%2C%20Madrid%2C%20Spain%2C%20Dec.%202021.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DEWUSPY69%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Network%20AI%20management%20%26%20orchestration%3A%20A%20federated%20multi-task%20learning%20case%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Wenliang%22%2C%22lastName%22%3A%22Liang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Chenghui%22%2C%22lastName%22%3A%22Peng%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xueli%22%2C%22lastName%22%3A%22An%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%226G%20treats%20artificial%20intelligence%20%28AI%29%20as%20the%20cornerstone%20and%20fundamental%20paradigm%20shift%20for%20providing%20inclusive%20intelligent%20services%2C%20which%20requires%20to%20natively%20support%20the%20training%20and%20reasoning%20of%20AI%20and%20provide%20a%20comprehensive%20network%20AI%20management%20%26%20orchestration%20%28NAMO%29%20solution.%20However%2C%20NAMO%20faces%20many%20practical%20challenges%20like%20multi-tenant%20multi-task%20coordination%2C%20heterogeneous%20resource%20scheduling%2C%20and%20security%20%26%20privacy%20concerns.%20In%20this%20paper%2C%20we%20take%20the%20federated%20multi-task%20learning%20as%20a%20starting%20case%20to%20demonstrate%20a%20promising%20NAMO%20solution.%20In%20particular%2C%20we%20propose%20a%20resource-aware%20method%20which%20leverages%20a%20primal-dual%20relationship%20to%20allow%20no%20direct%20uploading%20of%20local%20data%20to%20the%20edge%20server%20and%20maintain%20synchronous%20updates%20with%20straggler%20tolerance.%20Also%2C%20the%20proposed%20method%20could%20dynamically%20tune%20the%20learning%20accuracy%20at%20devices%20and%20the%20number%20of%20federated%20iterations%20to%20obtain%20a%20satisfactory%20training%20accuracy.%20Extensive%20simulation%20results%20have%20demonstrated%20the%20effectiveness%20of%20the%20proposed%20method.%22%2C%22date%22%3A%222021-12%22%2C%22proceedingsTitle%22%3A%22Proc.%20IEEE%20Globecom%202021%20%28EL5GMNB%20Workshop%29%22%2C%22conferenceName%22%3A%22IEEE%20Globecom%202021%20%28EL5GMNB%20Workshop%29%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222021-11-30T02%3A12%3A29Z%22%7D%7D%2C%7B%22key%22%3A%22XIFELVV7%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Liu%20et%20al.%22%2C%22parsedDate%22%3A%222021-11-01%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EZ.%20Liu%20%3Ci%3Eet%20al.%3C%5C%2Fi%3E%2C%20%26%23x201C%3BPersistent%20homology-based%20topological%20analysis%20on%20the%20gestalt%20patterns%20during%20human%20brain%20cognition%20process%2C%26%23x201D%3B%20%3Ci%3EJ.%20Healthcare%20Eng.%3C%5C%2Fi%3E%2C%20vol.%202021%2C%20p.%20e2334332%2C%20Nov.%202021.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DXIFELVV7%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Persistent%20homology-based%20topological%20analysis%20on%20the%20gestalt%20patterns%20during%20human%20brain%20cognition%20process%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zaisheng%22%2C%22lastName%22%3A%22Liu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Fei%22%2C%22lastName%22%3A%22Ni%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Chang%22%2C%22lastName%22%3A%22Liu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jiefang%22%2C%22lastName%22%3A%22Zhang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Songyun%22%2C%22lastName%22%3A%22Xie%22%7D%5D%2C%22abstractNote%22%3A%22The%20neuropsychological%20characteristics%20inside%20the%20brain%20are%20still%20not%20sufficiently%20understood%20in%20previous%20Gestalt%20psychological%20analyses.%20In%20particular%2C%20the%20extraction%20and%20analysis%20of%20human%20brain%20consciousness%20information%20itself%20have%20not%20received%20enough%20attention%20for%20the%20time%20being.%20In%20this%20paper%2C%20we%20aim%20to%20investigate%20the%20features%20of%20EEG%20signals%20from%20different%20conscious%20thoughts.%20Specifically%2C%20we%20try%20to%20extract%20the%20physiologically%20meaningful%20features%20of%20the%20brain%20responding%20to%20different%20contours%20and%20shapes%20in%20images%20in%20Gestalt%20cognitive%20tests%20by%20combining%20persistent%20homology%20analysis%20with%20electroencephalogram%20%28EEG%29.%20The%20experimental%20results%20show%20that%20more%20brain%20regions%20in%20the%20frontal%20lobe%20are%20involved%20when%20the%20subject%20perceives%20the%20random%20and%20disordered%20combination%20of%20images%20compared%20to%20the%20ordered%20Gestalt%20images.%20Meanwhile%2C%20the%20persistence%20entropy%20of%20EEG%20data%20evoked%20by%20random%20sequence%20diagram%20%28RSD%29%20is%20significantly%20different%20from%20that%20evoked%20by%20the%20ordered%20Gestalt%20%28GST%29%20images%20in%20several%20frequency%20bands%2C%20which%20indicate%20that%20the%20human%20cognition%20of%20the%20shape%20and%20contour%20of%20images%20can%20be%20separated%20to%20some%20extent%20through%20topological%20analysis.%20This%20implies%20the%20feasibility%20to%20digitize%20the%20neural%20signals%20while%20preserving%20the%20whole%20and%20local%20features%20of%20the%20original%20signals%2C%20which%20are%20further%20verified%20by%20our%20extensive%20experiments.%20In%20general%2C%20this%20paper%20evaluates%20and%20quantifies%20cognitively%20related%20neural%20correlates%20by%20persistent%20homology%20features%20of%20EEG%20signals%2C%20which%20provides%20an%20approach%20to%20realizing%20the%20digitization%20of%20neural%20signals.%20Preliminary%20verification%20of%20the%20analyzability%20of%20human%20consciousness%20signals%20provides%20reliable%20research%20ideas%20and%20directions%20for%20the%20realization%20of%20feature%20extraction%20and%20analysis%20of%20human%20brain%20consciousness%20cognition.%22%2C%22date%22%3A%222021%5C%2F11%5C%2F01%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222023-09-15T23%3A22%3A27Z%22%7D%7D%2C%7B%22key%22%3A%22T4D75WDM%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Shao%20et%20al.%22%2C%22parsedDate%22%3A%222021-10%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EY.%20Shao%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20B.%20Hu%2C%20Y.%20Wu%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%5C%22%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2021TVT.pdf%27%3EGraph%20attention%20network-based%20multi-agent%20reinforcement%20learning%20for%20slicing%20resource%20management%20in%20dense%20cellular%20network%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EIEEE%20Trans.%20Veh.%20Tech.%3C%5C%2Fi%3E%2C%20vol.%2070%2C%20no.%2010%2C%20pp.%2010792%5Cu201310803%2C%20Oct.%202021%2C%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DT4D75WDM%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Graph%20attention%20network-based%20multi-agent%20reinforcement%20learning%20for%20slicing%20resource%20management%20in%20dense%20cellular%20network%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yan%22%2C%22lastName%22%3A%22Shao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Bing%22%2C%22lastName%22%3A%22Hu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yingxiao%22%2C%22lastName%22%3A%22Wu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22Network%20slicing%20%28NS%29%20management%20devotes%20to%20provi-%20sioning%20various%20services%20with%20distinct%20requirements%20over%20the%20same%20physical%20communication%20infrastructure%20and%20allocating%20resources%20on%20demands.%20Considering%20a%20dense%20cellular%20network%20scenario%20that%20contains%20several%20NS%20over%20multiple%20base%20stations%20%28BSs%29%2C%20it%20remains%20challenging%20to%20design%20a%20proper%20resource%20management%20strategy%20in%20real%20time%2C%20so%20as%20to%20cope%20with%20frequent%20BS%20handover%20and%20meet%20distinct%20service%20requirements.%20In%20this%20paper%2C%20we%20propose%20to%20formulate%20this%20challenge%20as%20a%20multi-agent%20reinforcement%20learning%20%28MARL%29%20problem%20and%20leverage%20graph%20attention%20network%20%28GAT%29%20to%20strengthen%20the%20spatial%20cooperation%20between%20agents.%20Furthermore%2C%20we%20incorporate%20GAT%20into%20deep%20reinforcement%20learning%20%28DRL%29%20and%20correspondingly%20design%20an%20intelligent%20resource%20management%20strategy%20for%20NS.%20More%20specially%2C%20we%20have%20also%20testified%20the%20universal%20effectiveness%20of%20GAT%20for%20advancing%20DRL%20in%20multi-agent%20system%2C%20by%20applying%20GAT%20on%20the%20top%20of%20deep%20Q-network%20for%20the%20value-based%20method%20and%20advantage%20actor-critic%20for%20the%20policy-%20based%20method.%20Finally%2C%20we%20verify%20the%20superiority%20of%20the%20GAT-based%20DRL%20algorithms%20through%20extensive%20simulations.%22%2C%22date%22%3A%222021-10%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2021TVT.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222023-09-16T06%3A26%3A40Z%22%7D%7D%2C%7B%22key%22%3A%228UMVCMXC%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Chen%20et%20al.%22%2C%22parsedDate%22%3A%222021-06%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EY.%20Chen%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%5C%22%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2021TMC.pdf%27%3EOn%20the%20capacity%20of%20fractal%20D2D%20social%20networks%20with%20hierarchical%20communications%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EIEEE%20Trans.%20Mob.%20Comput.%3C%5C%2Fi%3E%2C%20vol.%2020%2C%20no.%206%2C%20pp.%202254%5Cu20132268%2C%20Jun.%202021%2C%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3D8UMVCMXC%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22On%20the%20capacity%20of%20fractal%20D2D%20social%20networks%20with%20hierarchical%20communications%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Ying%22%2C%22lastName%22%3A%22Chen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22The%20maximum%20capacity%20of%20fractal%20D2D%20%28device-to-device%29%20social%20networks%20with%20both%20direct%20and%20hierarchical%20communications%20is%20studied%20in%20this%20paper.%20Specifically%2C%20the%20fractal%20networks%20are%20characterized%20by%20the%20direct%20social%20connection%20and%20the%20self-similarity.%20Firstly%2C%20for%20a%20fractal%20D2D%20social%20network%20with%20direct%20social%20communications%2C%20it%20is%20proven%20that%20the%20maximum%20capacity%20is%20%24%5C%5CTheta%5C%5Cleft%28%5C%5Cfrac1%5C%5Csqrtn%5C%5Clog%20n%5C%5Cright%29%24%20if%20a%20user%20communicates%20with%20one%20of%20his%5C%2Fher%20direct%20contacts%20randomly%2C%20where%20n%20denotes%20the%20total%20number%20of%20users%20in%20the%20network%2C%20and%20it%20can%20reach%20up%20to%20%24%5C%5CTheta%5C%5Cleft%28%5C%5Cfrac1%5C%5Clog%20n%5C%5Cright%29%24%20if%20any%20pair%20of%20social%20contacts%20with%20distance%20d%20communicate%20according%20to%20the%20probability%20in%20proportion%20to%20%24d%5E-%5C%5Cbeta%24%2C%20where%20%24%5C%5Cbeta%24%20is%20the%20frequency%20parameter.%20Secondly%2C%20since%20users%20might%20get%20in%20touch%20with%20others%20without%20direct%20social%20connections%20through%20the%20inter-connected%20multiple%20users%2C%20the%20fractal%20D2D%20social%20network%20with%20these%20hierarchical%20communications%20is%20studied%20as%20well%2C%20and%20the%20related%20capacity%20is%20further%20derived.%20Our%20results%20show%20that%20this%20capacity%20is%20mainly%20affected%20by%20the%20correlation%20exponent%20%24%5C%5Cepsilon%24%20of%20the%20fractal%20structure.%20The%20capacity%20is%20reduced%20in%20proportional%20to%20%24%5C%5Cfrac1%5C%5Clog%20n%24%20if%20%242%5C%5Clt%20%5C%5Cepsilon%5C%5Clt%203%24%2C%20while%20the%20reduction%20coefficient%20is%20%24%5C%5Cfrac1n%24%20if%20%24%5C%5Cepsilon%5C%5Cgt%203%24.%22%2C%22date%22%3A%222021-6%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2021TMC.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222023-09-15T23%3A22%3A07Z%22%7D%7D%2C%7B%22key%22%3A%222PYW8696%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Pan%20et%20al.%22%2C%22parsedDate%22%3A%222021-06%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EJ.%20Pan%2C%20H.%20Shan%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Y.%20Wu%2C%20W.%20Wu%2C%20and%20T.%20Q.%20S.%20Quek%2C%20%26%23x201C%3BChannel%20estimation%20based%20on%20deep%20learning%20in%20vehicle-to-everything%20environments%2C%26%23x201D%3B%20%3Ci%3EIEEE%20Commun.%20Lett.%3C%5C%2Fi%3E%2C%20vol.%2025%2C%20no.%206%2C%20pp.%201891%26%23x2013%3B1895%2C%20Jun.%202021.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3D2PYW8696%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Channel%20estimation%20based%20on%20deep%20learning%20in%20vehicle-to-everything%20environments%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jing%22%2C%22lastName%22%3A%22Pan%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Hangguan%22%2C%22lastName%22%3A%22Shan%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yingxiao%22%2C%22lastName%22%3A%22Wu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Weihua%22%2C%22lastName%22%3A%22Wu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Tony%20Q.%20S.%22%2C%22lastName%22%3A%22Quek%22%7D%5D%2C%22abstractNote%22%3A%22Channel%20estimation%20in%20vehicle-to-everything%20%28V2X%29%20communications%20is%20a%20challenging%20issue%20due%20to%20the%20fast%20time-varying%20and%20non-stationary%20characteristics%20of%20wireless%20channel.%20To%20grasp%20the%20complicated%20variations%20of%20channel%20with%20limited%20number%20of%20pilots%20in%20the%20IEEE%20802.11p%20systems%2C%20data%20pilot-aided%20%28DPA%29%20channel%20estimation%20has%20been%20widely%20studied.%20However%2C%20the%20error%20propagation%20in%20the%20DPA%20procedure%2C%20caused%20by%20the%20noise%20and%20the%20channel%20variation%20within%20adjacent%20symbols%2C%20limits%20the%20performance%20seriously.%20In%20this%20letter%2C%20we%20propose%20a%20deep%20learning%20based%20channel%20estimation%20scheme%2C%20which%20exploits%20a%20long%20short-term%20memory%20network%20followed%20by%20a%20multilayer%20perceptron%20network%20to%20solve%20the%20error%20propagation%20issue.%20Simulation%20results%20show%20that%20the%20proposed%20scheme%20outperforms%20currently%20widely-used%20DPA%20schemes%20for%20the%20IEEE%20802.11p-based%20V2X%20communications.%22%2C%22date%22%3A%222021-6%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%5D%2C%22dateModified%22%3A%222023-09-16T06%3A26%3A42Z%22%7D%7D%2C%7B%22key%22%3A%22U4HPHKVC%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Chen%20et%20al.%22%2C%22parsedDate%22%3A%222021-06%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EM.%20Chen%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BRAN%20information-assisted%20TCP%20congestion%20control%20via%20DRL%20with%20reward%20redistribution%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20IEEE%20ICC%202021%20%28DDINS%20Workshop%29%3C%5C%2Fi%3E%2C%20Montreal%2C%20QC%2C%20Canada%2C%20Jun.%202021.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DU4HPHKVC%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22RAN%20information-assisted%20TCP%20congestion%20control%20via%20DRL%20with%20reward%20redistribution%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Minghao%22%2C%22lastName%22%3A%22Chen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%222021-6%22%2C%22proceedingsTitle%22%3A%22Proc.%20IEEE%20ICC%202021%20%28DDINS%20Workshop%29%22%2C%22conferenceName%22%3A%22IEEE%20ICC%202021%20%28DDINS%20Workshop%29%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222021-12-13T05%3A51%3A00Z%22%7D%7D%2C%7B%22key%22%3A%22AIYCLVEH%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Shao%20et%20al.%22%2C%22parsedDate%22%3A%222021-04%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EY.%20Shao%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BGraph%20attention%20network-based%20DRL%20for%20network%20slicing%20management%20in%20dense%20cellular%20networks%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20IEEE%20WCNC%202021%3C%5C%2Fi%3E%2C%20Nanjing%2C%20China%2C%20Apr.%202021.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DAIYCLVEH%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Graph%20attention%20network-based%20DRL%20for%20network%20slicing%20management%20in%20dense%20cellular%20networks%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yan%22%2C%22lastName%22%3A%22Shao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%222021-4%22%2C%22proceedingsTitle%22%3A%22Proc.%20IEEE%20WCNC%202021%22%2C%22conferenceName%22%3A%22IEEE%20WCNC%202021%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222021-01-07T06%3A56%3A23Z%22%7D%7D%2C%7B%22key%22%3A%22AI85SS74%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Shen%20et%20al.%22%2C%22parsedDate%22%3A%222021-01%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3ES.%20Shen%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20H.%20Zhang%2C%20and%20Y.%20Zhou%2C%20%26%23x201C%3BLearning%20to%20prune%20in%20training%20via%20dynamic%20channel%20propagation%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20ICPR%202021%3C%5C%2Fi%3E%2C%20Milan%2C%20Italy%2C%20Jan.%202021.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DAI85SS74%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Learning%20to%20prune%20in%20training%20via%20dynamic%20channel%20propagation%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Shibo%22%2C%22lastName%22%3A%22Shen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yugeng%22%2C%22lastName%22%3A%22Zhou%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%222021-1%22%2C%22proceedingsTitle%22%3A%22Proc.%20ICPR%202021%22%2C%22conferenceName%22%3A%22ICPR%202021%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222020-07-29T07%3A42%3A05Z%22%7D%7D%2C%7B%22key%22%3A%22E34DE7X5%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Shah%20et%20al.%22%2C%22parsedDate%22%3A%222021%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3ES.%20W.%20H.%20Shah%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20M.%20M.%20U.%20Rahman%2C%20A.%20N.%20Mian%2C%20W.%20Aman%2C%20and%20J.%20Crowcroft%2C%20%26%23x201C%3BStatistical%20QoS%20guarantees%20of%20a%20device-to-device%20link%20assisted%20by%20a%20full-duplex%20relay%2C%26%23x201D%3B%20%3Ci%3ETrans.%20Emerg.%20Telecommun.%20Technol.%20%28ETT%29%3C%5C%2Fi%3E%2C%202021.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DE34DE7X5%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Statistical%20QoS%20guarantees%20of%20a%20device-to-device%20link%20assisted%20by%20a%20full-duplex%20relay%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Syed%20Waqas%20Haider%22%2C%22lastName%22%3A%22Shah%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22M.%20Mahboob%20Ur%22%2C%22lastName%22%3A%22Rahman%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Adnan%20Noor%22%2C%22lastName%22%3A%22Mian%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Waqas%22%2C%22lastName%22%3A%22Aman%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jon%22%2C%22lastName%22%3A%22Crowcroft%22%7D%5D%2C%22abstractNote%22%3A%22Device-to-device%20%28D2D%29%20communication%20is%20a%20promising%20technique%20to%20enhance%20the%20spectral%20efficiency%20for%205G%20and%20beyond%20cellular%20networks.%20Traditionally%2C%20D2D%20communication%20was%20considered%20solely%20as%20communication%20directly%20between%20two%20closely-located%20devices%2C%20without%20any%20support%20from%20the%20available%20network%20infrastructure%3B%20however%2C%20this%20understanding%20limits%20the%20advantages%20of%20D2D%20communication.%20Relay-assisted%20D2D%20communication%20was%20later%20proposed%20to%20allow%20devices%20to%20exchange%20data%20and%20extend%20coverage%20via%20relay%20strategy%20for%20transmission%20distance%20extension%20and%20capacity%20improvements.%20This%20work%20aims%20to%20determine%20the%20statistical%20quality-of-service%20%28QoS%29%20guarantees%20of%20a%20D2D%20link%20assisted%20by%20a%20full-duplex%20relay%20by%20employing%20a%20commonly%20used%20analytical%20tool%2C%20the%20Effective%20capacity%20%28EC%29.%20We%20provide%20a%20closed-form%20expression%20for%20the%20EC%20of%20relay-assisted%20D2D%20communication%2C%20in%20which%20both%20the%20transmitter%20and%20the%20relay%20devices%20operate%20under%20individual%20QoS%20constraints.%20We%20also%20provide%20an%20analysis%20of%20the%20impact%20of%20mode%20selection%20on%20the%20EC%20of%20relay-assisted%20D2D.%20Additionally%2C%20for%20the%20mode%20selection%2C%20we%20propose%20a%20novel%20multiple%20features-based%20mechanism%20by%20utilizing%20all%20the%20features%20of%20a%20wireless%20link.%20We%20further%20provide%20a%20closed-form%20solution%20for%20the%20non-convex%20calibration%20problem%20of%20the%20optimal%20weights%20for%20these%20underlying%20features.%20Finally%2C%20simulation%20results%20demonstrate%20how%20the%20quality%20of%20self-interference%20cancellation%20techniques%20impacts%5Cnthe%20EC%20of%20the%20relay-assisted%20D2D.%20The%20results%20also%20show%20that%20the%20performance%20of%20our%20proposed%20mode%20selection%20mechanism%20with%20optimal%20weights%20outperforms%20the%20traditional%20mode%20selection%20mechanism.%22%2C%22date%22%3A%222021%22%2C%22language%22%3A%22en%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%5D%2C%22dateModified%22%3A%222023-09-16T06%3A26%3A40Z%22%7D%7D%2C%7B%22key%22%3A%223Q4TCRCW%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Shen%20et%20al.%22%2C%22parsedDate%22%3A%222020-12%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3ES.%20Shen%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20Q.%20Liu%2C%20J.%20Liang%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BEfficient%20deep%20structure%20learning%20for%20resource-limited%20IoT%20devices%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20IEEE%20Globecom%202020%3C%5C%2Fi%3E%2C%20Taipei%2C%20China%2C%20Dec.%202020.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3D3Q4TCRCW%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Efficient%20deep%20structure%20learning%20for%20resource-limited%20IoT%20devices%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Shibo%22%2C%22lastName%22%3A%22Shen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Qing%22%2C%22lastName%22%3A%22Liu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jing%22%2C%22lastName%22%3A%22Liang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%22Dec.%202020%22%2C%22proceedingsTitle%22%3A%22Proc.%20IEEE%20Globecom%202020%22%2C%22conferenceName%22%3A%22IEEE%20Globecom%202020%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222021-03-05T09%3A09%3A43Z%22%7D%7D%2C%7B%22key%22%3A%22G8QYVTJ5%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Wang%20et%20al.%22%2C%22parsedDate%22%3A%222020-11%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EG.%20Wang%2C%20Y.%20Zhong%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20X.%20Ge%2C%20T.%20Q.%20S.%20Quek%2C%20and%20G.%20Mao%2C%20%26%23x201C%3BEffect%20of%20spatial%20and%20temporal%20traffic%20statistics%20on%20the%20performance%20of%20wireless%20networks%2C%26%23x201D%3B%20%3Ci%3EIEEE%20Trans.%20Commun.%3C%5C%2Fi%3E%2C%20vol.%2068%2C%20no.%2011%2C%20pp.%207083%26%23x2013%3B7097%2C%20Nov.%202020.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DG8QYVTJ5%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Effect%20of%20spatial%20and%20temporal%20traffic%20statistics%20on%20the%20performance%20of%20wireless%20networks%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Gang%22%2C%22lastName%22%3A%22Wang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yi%22%2C%22lastName%22%3A%22Zhong%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xiaohu%22%2C%22lastName%22%3A%22Ge%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Tony%20Q.S.%22%2C%22lastName%22%3A%22Quek%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Guoqiang%22%2C%22lastName%22%3A%22Mao%22%7D%5D%2C%22abstractNote%22%3A%22The%20traffic%20in%20wireless%20networks%20has%20become%20diverse%20and%20fluctuating%20both%20spatially%20and%20temporally%20due%20to%20the%20emergence%20of%20new%20wireless%20applications%20and%20the%20complexity%20of%20scenarios.%20The%20purpose%20of%20this%20paper%20is%20to%20quantitatively%20analyze%20the%20impact%20of%20the%20wireless%20traffic%2C%20which%20fluctuates%20both%20spatially%20and%20temporally%2C%20on%20the%20performance%20of%20the%20wireless%20networks.%20Specially%2C%20we%20propose%20to%20combine%20the%20tools%20from%20stochastic%20geometry%20and%20queueing%20theory%20to%20model%20the%20spatial%20and%20temporal%20fluctuation%20of%20traffic%2C%20which%20to%20our%20best%20knowledge%20has%20seldom%20been%20evaluated%20analytically.%20We%20derive%20the%20spatial%20and%20temporal%20statistics%2C%20the%20total%20arrival%20rate%2C%20the%20stability%20of%20queues%20and%20the%20delay%20of%20users%20by%20considering%20two%20different%20spatial%20properties%20of%20traffic%2C%20i.e.%2C%20the%20uniformly%20and%20non-uniformly%20distributed%20cases.%20The%20numerical%20results%20indicate%20that%20although%20the%20fluctuation%20of%20traffic%20%28reflected%20by%20the%20variance%20of%20total%20arrival%20rate%29%20when%20the%20users%20are%20clustered%20is%20much%20fiercer%20than%20that%20when%20the%20users%20are%20uniformly%20distributed%2C%20the%20unstable%20probability%20is%20smaller.%20Our%20work%20provides%20a%20useful%20reference%20for%20the%20design%20of%20wireless%20networks%20when%20the%20complex%20spatio-temporal%20fluctuation%20of%20the%20traffic%20is%20considered.%22%2C%22date%22%3A%222020-11%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%221558-0857%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%5D%2C%22dateModified%22%3A%222023-09-15T23%3A26%3A16Z%22%7D%7D%2C%7B%22key%22%3A%22W2MPSNA7%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Dai%20et%20al.%22%2C%22parsedDate%22%3A%222020-10%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EA.%20Dai%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BGraph%20convolutional%20multi-agent%20reinforcement%20learning%20for%20uav%20coverage%20control%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20WCSP%202020%3C%5C%2Fi%3E%2C%20Nanjing%2C%20China%2C%20Oct.%202020.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DW2MPSNA7%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Graph%20convolutional%20multi-agent%20reinforcement%20learning%20for%20uav%20coverage%20control%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Anna%22%2C%22lastName%22%3A%22Dai%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%22October%202020%22%2C%22proceedingsTitle%22%3A%22Proc.%20WCSP%202020%22%2C%22conferenceName%22%3A%22WCSP%202020%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222021-01-07T06%3A56%3A11Z%22%7D%7D%2C%7B%22key%22%3A%22ZAIQ3GI5%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Chen%20et%20al.%22%2C%22parsedDate%22%3A%222020-10%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EK.%20Chen%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BThe%20implementation%20of%20asynchronous%20advantage%20actor-critic%20with%20stigmergy%20in%20network-assisted%20multi-agent%20system%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20WCSP%202020%3C%5C%2Fi%3E%2C%20Nanjing%2C%20China%2C%20Oct.%202020.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DZAIQ3GI5%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22The%20implementation%20of%20asynchronous%20advantage%20actor-critic%20with%20stigmergy%20in%20network-assisted%20multi-agent%20system%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Kun%22%2C%22lastName%22%3A%22Chen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%22October%202020%22%2C%22proceedingsTitle%22%3A%22Proc.%20WCSP%202020%22%2C%22conferenceName%22%3A%22WCSP%202020%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222021-01-07T06%3A55%3A45Z%22%7D%7D%2C%7B%22key%22%3A%228XFKVGKQ%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Su%20et%20al.%22%2C%22parsedDate%22%3A%222020-10%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EB.%20Su%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BEvolving%20deep%20convolutional%20neural%20network%20for%20intrusion%20detection%20based%20on%20neat%2C%26%23x201D%3B%20in%20%3Ci%3EProc.%20WPMC%202020%3C%5C%2Fi%3E%2C%20Virtual%20Edition%2C%20Oct.%202020.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3D8XFKVGKQ%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22Evolving%20deep%20convolutional%20neural%20network%20for%20intrusion%20detection%20based%20on%20neat%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Bingying%22%2C%22lastName%22%3A%22Su%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22To%20defend%20against%20the%20malicious%20attacks%20and%20the%20unauthorized%20activities%20in%20Internet%2C%20intrusion%20detection%20has%20been%20widely%20used%20to%20ensure%20network%20security.%20However%2C%20with%20continuously%20changing%20attacking%20methods%2C%20many%20challenges%20arise%20since%20the%20classical%20detection%20schemes%20fail%20to%20detect%20unknown%20attacks%20intelligently.%20Inspired%20by%20the%20recent%20success%20of%20deep%20learning%2C%20it%20is%20feasible%20to%20employ%20convolution%20neural%20network%20%28CNN%29%20to%20accomplish%20the%20intrusion%20detection%20task.%20Usually%20the%20performance%20of%20a%20CNN%20depends%20on%20finding%20an%20architecture%20to%20fit%20the%20task.%20In%20this%20paper%2C%20we%20propose%20an%20automated%20method%20for%20exploring%20an%20optimized%20CNN%20architecture%20which%20can%20effectively%20work%20on%20intrusion%20detection%20task%20by%20extending%20the%20Neuro%20Evolution%20of%20Augmenting%20Topologies%20%28NEAT%29%20algorithm.%20Evaluation%20results%20show%20that%20the%20proposed%20method%20is%20comparable%20to%20the%20best%20human%20designs%20regarding%20NSL-KDD%20datasets%20of%20intrusion%20detection.%22%2C%22date%22%3A%222020-10%22%2C%22proceedingsTitle%22%3A%22Proc.%20WPMC%202020%22%2C%22conferenceName%22%3A%22WPMC%202020%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222023-09-16T06%3A24%3A13Z%22%7D%7D%2C%7B%22key%22%3A%22GHID72S7%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Li%20et%20al.%22%2C%22parsedDate%22%3A%222020-09%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3E%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20C.%20Wang%2C%20R.%20Guo%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BThe%20LSTM-based%20advantage%20actor-critic%20learning%20for%20resource%20management%20in%20network%20slicing%20with%20user%20mobility%2C%26%23x201D%3B%20%3Ci%3EIEEE%20Commun.%20Lett.%3C%5C%2Fi%3E%2C%20vol.%2024%2C%20no.%209%2C%20pp.%202005%26%23x2013%3B2009%2C%20Sep.%202020.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DGHID72S7%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22The%20LSTM-based%20advantage%20actor-critic%20learning%20for%20resource%20management%20in%20network%20slicing%20with%20user%20mobility%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Chujie%22%2C%22lastName%22%3A%22Wang%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongbin%22%2C%22lastName%22%3A%22Guo%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22Network%20slicing%20aims%20to%20efficiently%20provision%20diversified%20services%20with%20distinct%20requirements%20over%20the%20same%20physical%20infrastructure.%20Therein%2C%20in%20order%20to%20efficiently%20allocate%20resources%20across%20slices%2C%20demand-aware%20inter-slice%20resource%20management%20is%20of%20significant%20importance.%20In%20this%20letter%2C%20we%20consider%20a%20scenario%20that%20contains%20several%20slices%20in%20a%20radio%20access%20network%20with%20base%20stations%20that%20share%20the%20same%20physical%20resources%20%28e.g.%2C%20bandwidth%20or%20slots%29.%20We%20primarily%20leverage%20advantage%20actor-critic%20%28A2C%29%2C%20one%20typical%20deep%20reinforcement%20learning%20%28DRL%29%20algorithm%2C%20to%20solve%20this%20problem%20by%20considering%20the%20varying%20service%20demands%20as%20the%20environment%20state%20and%20the%20allocated%20resources%20as%20the%20environment%20action.%20However%2C%20given%20that%20the%20user%20mobility%20toughens%20the%20difficulty%20to%20perceive%20the%20environment%2C%20we%20further%20incorporate%20the%20long%20short-term%20memory%20%28LSTM%29%20into%20A2C%2C%20and%20put%20forward%20an%20LSTM-A2C%20algorithm%20to%20track%20the%20user%20mobility%20and%20improve%20the%20system%20utility.%20We%20verify%20the%20performance%20of%20the%20proposed%20LSTM-A2C%20through%20extensive%20simulations.%22%2C%22date%22%3A%222020-9%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22IILGBZSM%22%2C%22B655FXSU%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222024-03-09T07%3A25%3A08Z%22%7D%7D%2C%7B%22key%22%3A%22P2TKB49L%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Chen%20et%20al.%22%2C%22parsedDate%22%3A%222020-09%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EK.%20Chen%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20J.%20Crowcroft%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%5C%22%3Ca%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2020MobicomDemo_Stigmergy.pdf%27%3EDEMO%3A%20The%20implementation%20of%20stigmergy%20in%20network-assisted%20multi-agent%20system%3C%5C%2Fa%3E%2C%5C%22%20in%20%3Ci%3EProc.%20ACM%20Mobicom%202020%3C%5C%2Fi%3E%2C%20London%2C%20UK%2C%20Sep.%202020.%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DP2TKB49L%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22conferencePaper%22%2C%22title%22%3A%22DEMO%3A%20The%20implementation%20of%20stigmergy%20in%20network-assisted%20multi-agent%20system%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Kun%22%2C%22lastName%22%3A%22Chen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Jon%22%2C%22lastName%22%3A%22Crowcroft%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22%22%2C%22date%22%3A%222020-9%22%2C%22proceedingsTitle%22%3A%22Proc.%20ACM%20Mobicom%202020%22%2C%22conferenceName%22%3A%22ACM%20Mobicom%202020%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISBN%22%3A%22%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_2020MobicomDemo_Stigmergy.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22TXJHVDQ3%22%5D%2C%22dateModified%22%3A%222022-02-18T03%3A24%3A42Z%22%7D%7D%2C%7B%22key%22%3A%22SFUKC3VK%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Li%20et%20al.%22%2C%22parsedDate%22%3A%222020-08%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3E%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20X.%20Xu%2C%20F.%20Ni%2C%20and%20H.%20Zhang%2C%20%26%23x201C%3BThe%20collective%20advantage%20for%20advancing%20communications%20and%20intelligence%2C%26%23x201D%3B%20%3Ci%3EIEEE%20Wireless%20Commun.%3C%5C%2Fi%3E%2C%20vol.%2027%2C%20no.%204%2C%20pp.%2096%26%23x2013%3B102%2C%20Aug.%202020.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DSFUKC3VK%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22The%20collective%20advantage%20for%20advancing%20communications%20and%20intelligence%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpeng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Xing%22%2C%22lastName%22%3A%22Xu%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Fei%22%2C%22lastName%22%3A%22Ni%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22The%20fifth-generation%20cellular%20networks%20%285G%29%20has%20boosted%20the%20unprecedented%5Cnconvergence%20between%20the%20information%20world%20and%20physical%20world.%20On%20the%20other%5Cnhand%2C%20empowered%20with%20the%20enormous%20amount%20of%20data%20and%20information%2C%20artificial%5Cnintelligence%20%28AI%29%20has%20been%20universally%20applied%20and%20pervasive%20AI%20is%20believed%20to%5Cnbe%20an%20integral%20part%20of%20the%20future%20cellular%20networks%20%28e.g.%2C%20beyond%205G%2C%20B5G%29.%5CnConsequently%2C%20benefiting%20from%20the%20advancement%20in%20communication%20technology%20and%5CnAI%2C%20we%20boldly%20argue%20that%20the%20conditions%20for%20collective%20intelligence%20%28CI%29%20will%5Cnbe%20mature%20in%20the%20B5G%20era%20and%20CI%20will%20emerge%20among%20the%20widely%20connected%20beings%5Cnand%20things.%20Afterwards%2C%20we%20highlight%20the%20potential%20huge%20impact%20of%20CI%20on%20both%5Cncommunications%20and%20intelligence.%20In%20particular%2C%20we%20introduce%20a%20regular%20language%5Cn%28i.e.%2C%20the%20information%20economy%20metalanguage%29%20supporting%20the%20future%20collective%5Cncommunications%20to%20augment%20human%20intelligence%20and%20explain%20its%20potential%5Cnapplications%20in%20naming%20Internet%20information%20and%20pushing%20information%20centric%5Cnnetworks%20forward.%20Meanwhile%2C%20we%20propose%20a%20stigmergy-based%20federated%20collective%5Cnintelligence%20and%20demonstrate%20its%20achievement%20in%20a%20simulated%20scenario%20where%20the%5Cnagents%20collectively%20work%20together%20to%20form%20a%20pattern%20through%20simple%20indirect%5Cncommunications.%20In%20a%20word%2C%20CI%20could%20advance%20both%20communications%20and%5Cnintelligence.%20discuss%20an%20anytime%20universal%20intelligence%20test%20model%20to%20evaluate%5Cnthe%20intelligence%20level%20of%20collective%20agents.%22%2C%22date%22%3A%222020-8%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%22IILGBZSM%22%2C%22RWWFPWPL%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222023-09-15T23%3A22%3A08Z%22%7D%7D%2C%7B%22key%22%3A%22CP92TYIQ%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Chen%20et%20al.%22%2C%22parsedDate%22%3A%222020-06%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EY.%20Chen%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20Z.%20Zhao%2C%20and%20H.%20Zhang%2C%20%5C%22%3Ca%20href%3D%27Https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_SystemJ2019Topology.pdf%27%3EStudy%20on%20base%20station%20topology%20in%20national%20cellular%20networks%3A%20Take%20advantage%20of%20alpha%20shapes%2C%20betti%20numbers%2C%20and%20euler%20characteristics%3C%5C%2Fa%3E%2C%5C%22%20%3Ci%3EIEEE%20Systems%20J.%3C%5C%2Fi%3E%2C%20vol.%2014%2C%20no.%202%2C%20pp.%202202%5Cu20132213%2C%20Jun.%202020%2C%20%5BOnline%5D.%20Available%3A%20%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DCP92TYIQ%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Study%20on%20base%20station%20topology%20in%20national%20cellular%20networks%3A%20Take%20advantage%20of%20alpha%20shapes%2C%20betti%20numbers%2C%20and%20euler%20characteristics%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Ying%22%2C%22lastName%22%3A%22Chen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rongpen%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhifeng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Honggang%22%2C%22lastName%22%3A%22Zhang%22%7D%5D%2C%22abstractNote%22%3A%22Faced%20with%20the%20ever-increasing%20trend%20of%20the%20cellular%20network%20scale%2C%20how%20to%20quantitatively%20evaluate%20the%20effectiveness%20of%20the%20large-scale%20deployment%20of%20base%20stations%20%28BSs%29%20has%20become%20a%20challenging%20topic.%20To%20this%20end%2C%20a%20deeper%20understanding%20of%20the%20cellular%20network%20topology%20is%20of%20fundamental%20significance%20to%20be%20achieved.%20In%20this%20paper%2C%20%24%20%5C%5Calpha%24-Shape%2C%20a%20powerful%20algebraic%20geometric%20tool%2C%20is%20integrated%20into%20the%20analyses%20of%20real%20BS%20location%20data%20for%20six%20Asian%20and%20six%20European%20countries.%20First%2C%20the%20BS%20spatial%20deployments%20of%20both%20Asian%20and%20European%20countries%20express%20fractal%20features%20based%20on%20two%20different%20testifying%20metrics%2C%20namely%20the%20Betti%20numbers%20and%20the%20Hurst%20coefficients.%20Second%2C%20it%20is%20found%20out%20that%20the%20log-normal%20distribution%20presents%20the%20best%20match%20to%20the%20cellular%20network%20topology%20when%20the%20practical%20BS%20deployment%20is%20characterized%20by%20the%20Euler%20characteristics.%22%2C%22date%22%3A%222020-6%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22Https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Ffiles%5C%2FPaper_SystemJ2019Topology.pdf%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%2C%22PMVJJHLU%22%5D%2C%22dateModified%22%3A%222023-09-15T23%3A22%3A08Z%22%7D%7D%2C%7B%22key%22%3A%22I2WDI44V%22%2C%22library%22%3A%7B%22id%22%3A1109573%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Shao%20et%20al.%22%2C%22parsedDate%22%3A%222020-05-01%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%3Cdiv%20class%3D%5C%22csl-bib-body%5C%22%20style%3D%5C%22line-height%3A%201.35%3B%20%5C%22%3E%5Cn%20%20%3Cdiv%20class%3D%5C%22csl-entry%5C%22%20style%3D%5C%22clear%3A%20left%3B%20%5C%22%3E%5Cn%20%20%20%20%3Cdiv%20class%3D%5C%22csl-left-margin%5C%22%20style%3D%5C%22float%3A%20left%3B%20padding-right%3A%200.5em%3B%20text-align%3A%20right%3B%20width%3A%201em%3B%5C%22%3E%5B1%5D%3C%5C%2Fdiv%3E%3Cdiv%20class%3D%5C%22csl-right-inline%5C%22%20style%3D%5C%22margin%3A%200%20.4em%200%201.5em%3B%5C%22%3EY.%20Shao%2C%20Z.%20Zhao%2C%20%3Cstrong%3ER.%20Li%3C%5C%2Fstrong%3E%2C%20and%20Y.%20Zhou%2C%20%26%23x201C%3BTarget%20detection%20for%20multi-UAVs%20via%20digital%20pheromones%20and%20navigation%20algorithm%20in%20unknown%20environments%2C%26%23x201D%3B%20%3Ci%3EFront.%20Inform.%20Technol.%20Electron.%20Eng.%3C%5C%2Fi%3E%2C%20vol.%2021%2C%20no.%205%2C%20pp.%20796%26%23x2013%3B808%2C%20May%202020.%20%3Ca%20title%3D%27Cite%20in%20RIS%20Format%27%20class%3D%27zp-CiteRIS%27%20href%3D%27https%3A%5C%2F%5C%2Fwww.rongpeng.info%5C%2Fli%5C%2Fwp-content%5C%2Fplugins%5C%2Fzotpress%5C%2Flib%5C%2Frequest%5C%2Frequest.cite.php%3Fapi_user_id%3D1109573%26amp%3Bitem_key%3DI2WDI44V%27%3ECite%3C%5C%2Fa%3E%20%3C%5C%2Fdiv%3E%5Cn%20%20%3C%5C%2Fdiv%3E%5Cn%3C%5C%2Fdiv%3E%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Target%20detection%20for%20multi-UAVs%20via%20digital%20pheromones%20and%20navigation%20algorithm%20in%20unknown%20environments%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yan%22%2C%22lastName%22%3A%22Shao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Zhi-feng%22%2C%22lastName%22%3A%22Zhao%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Rong-peng%22%2C%22lastName%22%3A%22Li%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yu-geng%22%2C%22lastName%22%3A%22Zhou%22%7D%5D%2C%22abstractNote%22%3A%22Coordinating%20multiple%20unmanned%20aerial%20vehicles%20%28multi-UAVs%29%20is%20a%20challenging%20technique%20in%20highly%20dynamic%20and%20sophisticated%20environments.%20Based%20on%20digital%20pheromones%20as%20well%20as%20current%20mainstream%20unmanned%20system%20controlling%20algorithms%2C%20we%20propose%20a%20strategy%20for%20multi-UAVs%20to%20acquire%20targets%20with%20limited%20prior%20knowledge.%20In%20particular%2C%20we%20put%20forward%20a%20more%20reasonable%20and%20effective%20pheromone%20update%20mechanism%2C%20by%20improving%20digital%20pheromone%20fusion%20algorithms%20for%20different%20semantic%20pheromones%20and%20planning%20individuals%5Cu2019%20probabilistic%20behavioral%20decision-making%20schemes.%20Also%2C%20inspired%20by%20the%20flocking%20model%20in%20nature%2C%20considering%20the%20limitations%20of%20some%20individuals%20in%20perception%20and%20communication%2C%20we%20design%20a%20navigation%20algorithm%20model%20on%20top%20of%20Olfati-Saber%5Cu2019s%20algorithm%20for%20flocking%20control%2C%20by%20further%20replacing%20the%20pheromone%20scalar%20to%20a%20vector.%20Simulation%20results%20show%20that%20the%20proposed%20algorithm%20can%20yield%20superior%20performance%20in%20terms%20of%20coverage%2C%20detection%20and%20revisit%20efficiency%2C%20and%20the%20capability%20of%20obstacle%20avoidance.%22%2C%22date%22%3A%222020-05-01%22%2C%22language%22%3A%22%22%2C%22DOI%22%3A%22%22%2C%22ISSN%22%3A%22%22%2C%22url%22%3A%22%22%2C%22collections%22%3A%5B%22XMUWZA6J%22%2C%224L54KHWH%22%5D%2C%22dateModified%22%3A%222023-09-15T23%3A26%3A14Z%22%7D%7D%5D%7D
[1]
Z. Lu, R. Li, X. Chen, E. Hossain, Z. Zhao, and H. Zhang, "Semantics-empowered communication: A tutorial-cum-survey," IEEE Commun. Surveys Tuts., vol. 26, no. 1, pp. 41–79, Mar. 2024, [Online]. Available: Cite
[1]
B. Xiao et al., "Stochastic graph neural network-based value decomposition for MARL in Internet of vehicles," IEEE Trans. Veh. Tech., vol. 73, no. 2, pp. 1582–1596, Feb. 2024, [Online]. Available: Cite
[1]
Y. Chen et al., “NetGPT: An AI-native network architecture for provisioning beyond personalized generative services,” IEEE Network, 2024. Cite
[1]
X. Xu, R. Li, Z. Zhao, and H. Zhang, "The gradient convergence bound of federated multi-agent reinforcement learning with efficient communication," IEEE Trans. Wireless Commun., vol. 23, no. 1, pp. 507–528, 2024, [Online]. Available: Cite
[1]
S. Li, Y. Xiang, R. Li, Z. Zhao, and H. Zhang, “Imitation learning based alternative multi-agent proximal policy optimization for well-formed swarm-oriented pursuit avoidance,” in Proc. ICCC 2023, Chengdu, China, Dec. 2023. Cite
[1]
X. Yu et al., “Communication-efficient cooperative multi-agent PPO via regulated segment mixture in internet of vehicles,” in Proc. IEEE Globecom 2023, Kuala Lumpur, Malaysia, Dec. 2023. Cite
[1]
Y. Xiang, S. Li, R. Li, Z. Zhao, and H. Zhang, “Decentralized adaptive formation via consensus-oriented multi-agent communication,” in Proc. WCSP 2023, Hangzhou, China, Nov. 2023. Cite
[1]
W. Geng, B. Xiao, R. Li, N. Wei, Z. Zhao, and H. Zhang, “Decomposition-based multi-agent distributional reinforcement learning for task-oriented UAV collaboration with noisy rewards,” in Proc. WCSP 2023, Hangzhou, China, Nov. 2023. Cite
[1]
F. Ni, Z. Zhao, R. Li, and H. Zhang, "EEG signal-assisted algebraic topological feature-enhanced deep neural networks for gestalt illusory contour perception," IEEE Access, vol. 11, pp. 96029–96042, Sep. 2023, [Online]. Available: Cite
[1]
S. Tong, X. Yu, R. Li, K. Lu, Z. Zhao, and H. Zhang, “Alternate learning based sparse semantic communications for visual transmission,” in Proc. IEEE PIMRC 2023, Toronto, Canada, Sep. 2023. Cite
[1]
Y. Liu, D. Li, R. Li, Z. Zhao, Y. Zhu, and H. Zhang, "Secure and efficient stigmergy-empowered blockchain framework for heterogeneous collaborative services in the Internet of vehicles," IEEE Commun. Mag., vol. 61, no. 9, pp. 186-- 192, Sep. 2023, [Online]. Available: Cite
[1]
B. Wang, R. Li, J. Zhu, Z. Zhao, and H. Zhang, “Knowledge enhanced semantic communication receiver,” IEEE Commun. Lett., vol. 27, no. 7, pp. 1794–1798, Jul. 2023. Cite
[1]
J. Zhu, R. Li, X. Chen, S. Mao, J. Wu, and Z. Zhao, “Semantics-enhanced temporal graph networks for content caching and energy saving,” in Proc. IEEE ICC 2023, Rome, Italy, Jun. 2023. Cite
[1]
J. Fang, C. Wang, R. Li, H. Wei, and M. Zhao, “Matrix factorization and deep autoencoder based clustering scheme for large-scale UAV networks,” in Proc. IEEE VTC 2023-Spring, Florence, Italy, Jun. 2023. Cite
[1]
B. Xiao et al., “Stochastic graph neural network-based value decomposition for multi-agent reinforcement learning in urban traffic control,” in Proc. IEEE VTC 2023-Spring, Florence, Italy, Jun. 2023. Cite
[1]
J. Zhu et al., "AoI-based temporal attention graph neural network for popularity prediction and content caching," IEEE Trans. Cog. Commun. Netw., vol. 9, no. 2, pp. 345–358, Apr. 2023, [Online]. Available: Cite
[1]
K. Lu et al., "Rethinking modern communication from semantic coding to semantic communication," IEEE Wireless Commun., vol. 30, no. 1, pp. 158–164, Feb. 2023, [Online]. Available: Cite
[1]
J. Zhu, R. Li, X. Chen, S. Mao, J. Wu, and Z. Zhao, “Semantics-enhanced temporal graph networks for content popularity prediction,” IEEE Trans. Mob. Comput., 2023, doi: 10.1109/TMC.2023.3349315. Cite
[1]
Y. Yang et al., “Task-oriented 6G native-AI network architecture,” IEEE Network, pp. 1–1, 2023, doi: 10.1109/MNET.2023.3321464. Cite
[1]
H. Wei, C. Wang, R. Li, and M. Zhao, “Mean-field MARL-based priority-aware CSMA/CA strategy in large-scale manets,” in Proc. IEEE Globecom 2022, Rio de Janeiro, Brazil, Dec. 2022. Cite
[1]
K. Lu, R. Li, and H. Zhang, “Contrastive monotonic pixel-level modulation,” in Proc. ECCV, in ECCV ’22. Tel-Aviv, Oct. 2022. Cite
[1]
X. Xu, R. Li, Z. Zhao, and H. Zhang, "Stigmergic independent reinforcement learning for multiagent collaboration," IEEE Trans. Neural Netw. Learn. Syst., vol. 33, no. 9, pp. 4285–4299, Sep. 2022, [Online]. Available: Cite
[1]
X. Yi, R. Li, C. Peng, J. Wu, and Z. Zhao, “HFedMTL: Hierarchical federated multi-task learning,” in Proc. IEEE PIMRC 2022, Virtual Edition, Sep. 2022. Cite
[1]
Q. Zhou, R. Li, Z. Zhao, Y. Xiao, and H. Zhang, "Adaptive bit rate control in semantic communication with incremental knowledge-based HARQ," IEEE Open J. Commun. Soc., vol. 3, pp. 1076–1089, Jul. 2022, [Online]. Available: Cite
[1]
X. Xu, R. Li, Z. Zhao, and H. Zhang, “Communication-efficient consensus mechanism for federated reinforcement learning,” in Proc. IEEE ICC 2022, Seoul, South Korea, May 2022. Cite
[1]
B. Xiao, Y. Shao, R. Li, Z. Zhao, and H. Zhang, "DEMO: Deep reinforcement learning for resource management in cellular network slicing," in Proc. IEEE INFOCOM 2022, in INFOCOM '22. Virtual Edition, May 2022. [Online]. Available: Cite
[1]
X. Xu, R. Li, Z. Zhao, and H. Zhang, "Trustable policy collaboration scheme for multi-agent stigmergic reinforcement learning," IEEE Commun. Lett., vol. 26, no. 4, pp. 823–827, Apr. 2022, [Online]. Available: Cite
[1]
J. Zhu, R. Li, Z. Zhao, and H. Zhang, “AoI-based temporal attention graph neural network for popularity prediction in ICN,” in Proc. IEEE WCNC 2022, in WCNC ’22. Austin, TX, USA, Apr. 2022. Cite
[1]
Q. Zhou, R. Li, Z. Zhao, C. Peng, and H. Zhang, "Semantic communication with adaptive universal transformer," IEEE Wirel. Commun. Lett., vol. 11, no. 3, pp. 453–457, Mar. 2022, [Online]. Available: Cite
[1]
M. Chen, R. Li, J. Crowcroft, J. Wu, Z. Zhao, and H. Zhang, "RAN information-assisted TCP congestion control using deep reinforcement learning with reward redistribution," IEEE Trans. Commun., vol. 70, no. 1, pp. 215–230, Jan. 2022, [Online]. Available: Cite
[1]
C. Qi et al., "An efficient pruning scheme of deep neural networks for Internet of things applications," EURASIP J. Adv. Signal Process., vol. 2021, no. 1, pp. 1–21, Dec. 2021, Accessed: Nov. 10, 2021. [Online]. Available: Cite
[1]
R. Li, W. Liang, C. Peng, X. An, Z. Zhao, and H. Zhang, “Network AI management & orchestration: A federated multi-task learning case,” in Proc. IEEE Globecom 2021 (EL5GMNB Workshop), Madrid, Spain, Dec. 2021. Cite
[1]
Z. Liu et al., “Persistent homology-based topological analysis on the gestalt patterns during human brain cognition process,” J. Healthcare Eng., vol. 2021, p. e2334332, Nov. 2021. Cite
[1]
Y. Shao, R. Li, B. Hu, Y. Wu, Z. Zhao, and H. Zhang, "Graph attention network-based multi-agent reinforcement learning for slicing resource management in dense cellular network," IEEE Trans. Veh. Tech., vol. 70, no. 10, pp. 10792–10803, Oct. 2021, [Online]. Available: Cite
[1]
Y. Chen, R. Li, Z. Zhao, and H. Zhang, "On the capacity of fractal D2D social networks with hierarchical communications," IEEE Trans. Mob. Comput., vol. 20, no. 6, pp. 2254–2268, Jun. 2021, [Online]. Available: Cite
[1]
J. Pan, H. Shan, R. Li, Y. Wu, W. Wu, and T. Q. S. Quek, “Channel estimation based on deep learning in vehicle-to-everything environments,” IEEE Commun. Lett., vol. 25, no. 6, pp. 1891–1895, Jun. 2021. Cite
[1]
M. Chen, R. Li, Z. Zhao, and H. Zhang, “RAN information-assisted TCP congestion control via DRL with reward redistribution,” in Proc. IEEE ICC 2021 (DDINS Workshop), Montreal, QC, Canada, Jun. 2021. Cite
[1]
Y. Shao, R. Li, Z. Zhao, and H. Zhang, “Graph attention network-based DRL for network slicing management in dense cellular networks,” in Proc. IEEE WCNC 2021, Nanjing, China, Apr. 2021. Cite
[1]
S. Shen, R. Li, Z. Zhao, H. Zhang, and Y. Zhou, “Learning to prune in training via dynamic channel propagation,” in Proc. ICPR 2021, Milan, Italy, Jan. 2021. Cite
[1]
S. W. H. Shah, R. Li, M. M. U. Rahman, A. N. Mian, W. Aman, and J. Crowcroft, “Statistical QoS guarantees of a device-to-device link assisted by a full-duplex relay,” Trans. Emerg. Telecommun. Technol. (ETT), 2021. Cite
[1]
S. Shen, R. Li, Z. Zhao, Q. Liu, J. Liang, and H. Zhang, “Efficient deep structure learning for resource-limited IoT devices,” in Proc. IEEE Globecom 2020, Taipei, China, Dec. 2020. Cite
[1]
G. Wang, Y. Zhong, R. Li, X. Ge, T. Q. S. Quek, and G. Mao, “Effect of spatial and temporal traffic statistics on the performance of wireless networks,” IEEE Trans. Commun., vol. 68, no. 11, pp. 7083–7097, Nov. 2020. Cite
[1]
A. Dai, R. Li, Z. Zhao, and H. Zhang, “Graph convolutional multi-agent reinforcement learning for uav coverage control,” in Proc. WCSP 2020, Nanjing, China, Oct. 2020. Cite
[1]
K. Chen, R. Li, Z. Zhao, and H. Zhang, “The implementation of asynchronous advantage actor-critic with stigmergy in network-assisted multi-agent system,” in Proc. WCSP 2020, Nanjing, China, Oct. 2020. Cite
[1]
B. Su, R. Li, and H. Zhang, “Evolving deep convolutional neural network for intrusion detection based on neat,” in Proc. WPMC 2020, Virtual Edition, Oct. 2020. Cite
[1]
R. Li, C. Wang, R. Guo, Z. Zhao, and H. Zhang, “The LSTM-based advantage actor-critic learning for resource management in network slicing with user mobility,” IEEE Commun. Lett., vol. 24, no. 9, pp. 2005–2009, Sep. 2020. Cite
[1]
K. Chen, R. Li, J. Crowcroft, Z. Zhao, and H. Zhang, "DEMO: The implementation of stigmergy in network-assisted multi-agent system," in Proc. ACM Mobicom 2020, London, UK, Sep. 2020. [Online]. Available: Cite
[1]
R. Li, Z. Zhao, X. Xu, F. Ni, and H. Zhang, “The collective advantage for advancing communications and intelligence,” IEEE Wireless Commun., vol. 27, no. 4, pp. 96–102, Aug. 2020. Cite
[1]
Y. Chen, R. Li, Z. Zhao, and H. Zhang, "Study on base station topology in national cellular networks: Take advantage of alpha shapes, betti numbers, and euler characteristics," IEEE Systems J., vol. 14, no. 2, pp. 2202–2213, Jun. 2020, [Online]. Available: Cite
[1]
Y. Shao, Z. Zhao, R. Li, and Y. Zhou, “Target detection for multi-UAVs via digital pheromones and navigation algorithm in unknown environments,” Front. Inform. Technol. Electron. Eng., vol. 21, no. 5, pp. 796–808, May 2020. Cite