Matches in SemOpenAlex for { <https://semopenalex.org/work/W4380078504> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W4380078504 endingPage "1" @default.
- W4380078504 startingPage "1" @default.
- W4380078504 abstract "The paper considers independent reinforcement learning (IRL) for multi-agent collaborative decision-making in the paradigm of federated learning (FL). However, FL generates excessive communication overheads between agents and a remote central server, especially when it involves a large number of agents or iterations. Besides, due to the heterogeneity of independent learning environments, multiple agents may undergo asynchronous Markov decision processes (MDPs), which will affect the training samples and the model’s convergence performance. On top of the variation-aware periodic averaging (VPA) method and the policy-based deep reinforcement learning (DRL) algorithm (i.e., proximal policy optimization (PPO)), this paper proposes two advanced optimization schemes orienting to stochastic gradient descent (SGD): 1) A decay-based scheme gradually decays the weights of a model’s local gradients with the progress of successive local updates, and 2) By representing the agents as a graph, a consensus-based scheme studies the impact of exchanging a model’s local gradients among nearby agents from an algebraic connectivity perspective. This paper also provides novel convergence guarantees for both developed schemes, and demonstrates their superior effectiveness and efficiency in improving the system’s utility value through theoretical analyses and simulation results." @default.
- W4380078504 created "2023-06-10" @default.
- W4380078504 creator A5013688157 @default.
- W4380078504 creator A5027890741 @default.
- W4380078504 creator A5034572686 @default.
- W4380078504 creator A5036831868 @default.
- W4380078504 date "2023-01-01" @default.
- W4380078504 modified "2023-09-27" @default.
- W4380078504 title "The Gradient Convergence Bound of Federated Multi-Agent Reinforcement Learning with Efficient Communication" @default.
- W4380078504 doi "https://doi.org/10.1109/twc.2023.3279268" @default.
- W4380078504 hasPublicationYear "2023" @default.
- W4380078504 type Work @default.
- W4380078504 citedByCount "0" @default.
- W4380078504 crossrefType "journal-article" @default.
- W4380078504 hasAuthorship W4380078504A5013688157 @default.
- W4380078504 hasAuthorship W4380078504A5027890741 @default.
- W4380078504 hasAuthorship W4380078504A5034572686 @default.
- W4380078504 hasAuthorship W4380078504A5036831868 @default.
- W4380078504 hasConcept C105795698 @default.
- W4380078504 hasConcept C106189395 @default.
- W4380078504 hasConcept C126255220 @default.
- W4380078504 hasConcept C12713177 @default.
- W4380078504 hasConcept C134306372 @default.
- W4380078504 hasConcept C151319957 @default.
- W4380078504 hasConcept C153258448 @default.
- W4380078504 hasConcept C154945302 @default.
- W4380078504 hasConcept C159886148 @default.
- W4380078504 hasConcept C162324750 @default.
- W4380078504 hasConcept C206688291 @default.
- W4380078504 hasConcept C2777303404 @default.
- W4380078504 hasConcept C31258907 @default.
- W4380078504 hasConcept C33923547 @default.
- W4380078504 hasConcept C41008148 @default.
- W4380078504 hasConcept C50522688 @default.
- W4380078504 hasConcept C50644808 @default.
- W4380078504 hasConcept C77618280 @default.
- W4380078504 hasConcept C97541855 @default.
- W4380078504 hasConceptScore W4380078504C105795698 @default.
- W4380078504 hasConceptScore W4380078504C106189395 @default.
- W4380078504 hasConceptScore W4380078504C126255220 @default.
- W4380078504 hasConceptScore W4380078504C12713177 @default.
- W4380078504 hasConceptScore W4380078504C134306372 @default.
- W4380078504 hasConceptScore W4380078504C151319957 @default.
- W4380078504 hasConceptScore W4380078504C153258448 @default.
- W4380078504 hasConceptScore W4380078504C154945302 @default.
- W4380078504 hasConceptScore W4380078504C159886148 @default.
- W4380078504 hasConceptScore W4380078504C162324750 @default.
- W4380078504 hasConceptScore W4380078504C206688291 @default.
- W4380078504 hasConceptScore W4380078504C2777303404 @default.
- W4380078504 hasConceptScore W4380078504C31258907 @default.
- W4380078504 hasConceptScore W4380078504C33923547 @default.
- W4380078504 hasConceptScore W4380078504C41008148 @default.
- W4380078504 hasConceptScore W4380078504C50522688 @default.
- W4380078504 hasConceptScore W4380078504C50644808 @default.
- W4380078504 hasConceptScore W4380078504C77618280 @default.
- W4380078504 hasConceptScore W4380078504C97541855 @default.
- W4380078504 hasLocation W43800785041 @default.
- W4380078504 hasOpenAccess W4380078504 @default.
- W4380078504 hasPrimaryLocation W43800785041 @default.
- W4380078504 hasRelatedWork W1556532828 @default.
- W4380078504 hasRelatedWork W1574991376 @default.
- W4380078504 hasRelatedWork W1626977535 @default.
- W4380078504 hasRelatedWork W1985560493 @default.
- W4380078504 hasRelatedWork W2145363145 @default.
- W4380078504 hasRelatedWork W2937181779 @default.
- W4380078504 hasRelatedWork W3139092397 @default.
- W4380078504 hasRelatedWork W3213537191 @default.
- W4380078504 hasRelatedWork W4287868411 @default.
- W4380078504 hasRelatedWork W4386148312 @default.
- W4380078504 isParatext "false" @default.
- W4380078504 isRetracted "false" @default.
- W4380078504 workType "article" @default.