Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313291205> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W4313291205 endingPage "13" @default.
- W4313291205 startingPage "1" @default.
- W4313291205 abstract "Multi-agent settings remain a fundamental challenge in the reinforcement learning (RL) domain due to the partial observability and the lack of accurate real-time interactions across agents. In this paper, we propose a new method based on local communication learning to tackle the multi-agent RL (MARL) challenge within a large number of agents coexisting. First, we design a new communication protocol that exploits the ability of depthwise convolution to efficiently extract local relations and learn local communication between neighboring agents. To facilitate multi-agent coordination, we explicitly learn the effect of joint actions by taking the policies of neighboring agents as inputs. Second, we introduce the mean-field approximation into our method to reduce the scale of agent interactions. To more effectively coordinate behaviors of neighboring agents, we enhance the mean-field approximation by a supervised policy rectification network (PRN) for rectifying real-time agent interactions and by a learnable compensation term for correcting the approximation bias. The proposed method enables efficient coordination as well as outperforms several baseline approaches on the adaptive traffic signal control (ATSC) task and the StarCraft II multi-agent challenge (SMAC)." @default.
- W4313291205 created "2023-01-06" @default.
- W4313291205 creator A5003514239 @default.
- W4313291205 creator A5035185341 @default.
- W4313291205 creator A5036016529 @default.
- W4313291205 creator A5047526614 @default.
- W4313291205 date "2022-01-01" @default.
- W4313291205 modified "2023-09-25" @default.
- W4313291205 title "Depthwise Convolution for Multi-Agent Communication With Enhanced Mean-Field Approximation" @default.
- W4313291205 doi "https://doi.org/10.1109/tnnls.2022.3230701" @default.
- W4313291205 hasPublicationYear "2022" @default.
- W4313291205 type Work @default.
- W4313291205 citedByCount "0" @default.
- W4313291205 crossrefType "journal-article" @default.
- W4313291205 hasAuthorship W4313291205A5003514239 @default.
- W4313291205 hasAuthorship W4313291205A5035185341 @default.
- W4313291205 hasAuthorship W4313291205A5036016529 @default.
- W4313291205 hasAuthorship W4313291205A5047526614 @default.
- W4313291205 hasBestOaLocation W43132912052 @default.
- W4313291205 hasConcept C120314980 @default.
- W4313291205 hasConcept C134306372 @default.
- W4313291205 hasConcept C142724271 @default.
- W4313291205 hasConcept C154945302 @default.
- W4313291205 hasConcept C204787440 @default.
- W4313291205 hasConcept C2780385302 @default.
- W4313291205 hasConcept C28826006 @default.
- W4313291205 hasConcept C33923547 @default.
- W4313291205 hasConcept C36299963 @default.
- W4313291205 hasConcept C36503486 @default.
- W4313291205 hasConcept C41008148 @default.
- W4313291205 hasConcept C41550386 @default.
- W4313291205 hasConcept C45347329 @default.
- W4313291205 hasConcept C50644808 @default.
- W4313291205 hasConcept C71924100 @default.
- W4313291205 hasConcept C97541855 @default.
- W4313291205 hasConceptScore W4313291205C120314980 @default.
- W4313291205 hasConceptScore W4313291205C134306372 @default.
- W4313291205 hasConceptScore W4313291205C142724271 @default.
- W4313291205 hasConceptScore W4313291205C154945302 @default.
- W4313291205 hasConceptScore W4313291205C204787440 @default.
- W4313291205 hasConceptScore W4313291205C2780385302 @default.
- W4313291205 hasConceptScore W4313291205C28826006 @default.
- W4313291205 hasConceptScore W4313291205C33923547 @default.
- W4313291205 hasConceptScore W4313291205C36299963 @default.
- W4313291205 hasConceptScore W4313291205C36503486 @default.
- W4313291205 hasConceptScore W4313291205C41008148 @default.
- W4313291205 hasConceptScore W4313291205C41550386 @default.
- W4313291205 hasConceptScore W4313291205C45347329 @default.
- W4313291205 hasConceptScore W4313291205C50644808 @default.
- W4313291205 hasConceptScore W4313291205C71924100 @default.
- W4313291205 hasConceptScore W4313291205C97541855 @default.
- W4313291205 hasLocation W43132912051 @default.
- W4313291205 hasLocation W43132912052 @default.
- W4313291205 hasLocation W43132912053 @default.
- W4313291205 hasOpenAccess W4313291205 @default.
- W4313291205 hasPrimaryLocation W43132912051 @default.
- W4313291205 hasRelatedWork W1481751858 @default.
- W4313291205 hasRelatedWork W2002084071 @default.
- W4313291205 hasRelatedWork W2020274357 @default.
- W4313291205 hasRelatedWork W2120073855 @default.
- W4313291205 hasRelatedWork W2121901516 @default.
- W4313291205 hasRelatedWork W2127038482 @default.
- W4313291205 hasRelatedWork W2166189192 @default.
- W4313291205 hasRelatedWork W3201419494 @default.
- W4313291205 hasRelatedWork W4312858357 @default.
- W4313291205 hasRelatedWork W4313291432 @default.
- W4313291205 isParatext "false" @default.
- W4313291205 isRetracted "false" @default.
- W4313291205 workType "article" @default.