Matches in SemOpenAlex for { <https://semopenalex.org/work/W4378446622> ?p ?o ?g. }
Showing items 1 to 63 of
63
with 100 items per page.
- W4378446622 abstract "Cooperative Multi-agent Reinforcement Learning (MARL) has attracted significant attention and played the potential for many real-world applications. Previous arts mainly focus on facilitating the coordination ability from different aspects (e.g., non-stationarity, credit assignment) in single-task or multi-task scenarios, ignoring the stream of tasks that appear in a continual manner. This ignorance makes the continual coordination an unexplored territory, neither in problem formulation nor efficient algorithms designed. Towards tackling the mentioned issue, this paper proposes an approach Multi-Agent Continual Coordination via Progressive Task Contextualization, dubbed MACPro. The key point lies in obtaining a factorized policy, using shared feature extraction layers but separated independent task heads, each specializing in a specific class of tasks. The task heads can be progressively expanded based on the learned task contextualization. Moreover, to cater to the popular CTDE paradigm in MARL, each agent learns to predict and adopt the most relevant policy head based on local information in a decentralized manner. We show in multiple multi-agent benchmarks that existing continual learning methods fail, while MACPro is able to achieve close-to-optimal performance. More results also disclose the effectiveness of MACPro from multiple aspects like high generalization ability." @default.
- W4378446622 created "2023-05-27" @default.
- W4378446622 creator A5033612713 @default.
- W4378446622 creator A5046214153 @default.
- W4378446622 creator A5088385429 @default.
- W4378446622 creator A5088477939 @default.
- W4378446622 creator A5089801879 @default.
- W4378446622 creator A5092028523 @default.
- W4378446622 date "2023-05-07" @default.
- W4378446622 modified "2023-10-16" @default.
- W4378446622 title "Multi-agent Continual Coordination via Progressive Task Contextualization" @default.
- W4378446622 doi "https://doi.org/10.48550/arxiv.2305.13937" @default.
- W4378446622 hasPublicationYear "2023" @default.
- W4378446622 type Work @default.
- W4378446622 citedByCount "0" @default.
- W4378446622 crossrefType "posted-content" @default.
- W4378446622 hasAuthorship W4378446622A5033612713 @default.
- W4378446622 hasAuthorship W4378446622A5046214153 @default.
- W4378446622 hasAuthorship W4378446622A5088385429 @default.
- W4378446622 hasAuthorship W4378446622A5088477939 @default.
- W4378446622 hasAuthorship W4378446622A5089801879 @default.
- W4378446622 hasAuthorship W4378446622A5092028523 @default.
- W4378446622 hasBestOaLocation W43784466221 @default.
- W4378446622 hasConcept C107457646 @default.
- W4378446622 hasConcept C127413603 @default.
- W4378446622 hasConcept C154945302 @default.
- W4378446622 hasConcept C199360897 @default.
- W4378446622 hasConcept C201995342 @default.
- W4378446622 hasConcept C26517878 @default.
- W4378446622 hasConcept C2780451532 @default.
- W4378446622 hasConcept C2780712339 @default.
- W4378446622 hasConcept C38652104 @default.
- W4378446622 hasConcept C41008148 @default.
- W4378446622 hasConcept C527412718 @default.
- W4378446622 hasConcept C97541855 @default.
- W4378446622 hasConceptScore W4378446622C107457646 @default.
- W4378446622 hasConceptScore W4378446622C127413603 @default.
- W4378446622 hasConceptScore W4378446622C154945302 @default.
- W4378446622 hasConceptScore W4378446622C199360897 @default.
- W4378446622 hasConceptScore W4378446622C201995342 @default.
- W4378446622 hasConceptScore W4378446622C26517878 @default.
- W4378446622 hasConceptScore W4378446622C2780451532 @default.
- W4378446622 hasConceptScore W4378446622C2780712339 @default.
- W4378446622 hasConceptScore W4378446622C38652104 @default.
- W4378446622 hasConceptScore W4378446622C41008148 @default.
- W4378446622 hasConceptScore W4378446622C527412718 @default.
- W4378446622 hasConceptScore W4378446622C97541855 @default.
- W4378446622 hasLocation W43784466221 @default.
- W4378446622 hasOpenAccess W4378446622 @default.
- W4378446622 hasPrimaryLocation W43784466221 @default.
- W4378446622 hasRelatedWork W1562959674 @default.
- W4378446622 hasRelatedWork W2613109121 @default.
- W4378446622 hasRelatedWork W2774891019 @default.
- W4378446622 hasRelatedWork W2923653485 @default.
- W4378446622 hasRelatedWork W2952472710 @default.
- W4378446622 hasRelatedWork W2957776456 @default.
- W4378446622 hasRelatedWork W3209094908 @default.
- W4378446622 hasRelatedWork W4210912933 @default.
- W4378446622 hasRelatedWork W4287626175 @default.
- W4378446622 hasRelatedWork W4361026739 @default.
- W4378446622 isParatext "false" @default.
- W4378446622 isRetracted "false" @default.
- W4378446622 workType "article" @default.