Matches in SemOpenAlex for { <https://semopenalex.org/work/W4381164013> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W4381164013 endingPage "22" @default.
- W4381164013 startingPage "1" @default.
- W4381164013 abstract "Federated Learning (FL) learns a global model in a distributional manner, which does not require local clients to share private data. Such merit has drawn lots of attention in the interaction scenarios, where Federated Reinforcement Learning (FRL) emerges as a cross-field research direction focusing on the robust training of agents. Different from FL, the heterogeneity problem in FRL is more challenging because the data depends on the policy of agents and the environment dynamics. FRL learns to interact under the non-stationary environment feedback, while the typical FL methods aim at handling the constant data heterogeneity. In this article, we are among the first attempts to analyze the heterogeneity problem in FRL and propose an off-policy FRL framework. Specifically, a student–teacher–student model learning and fusion method, termed as Server-Client Collaborative Distillation (SCCD), is introduced. Unlike the traditional FL, we distill all local models on the server side for model fusion. To reduce the variance of the training, a local distillation is also conducted every time the agent receives the global model. Experimentally, we compare SCCD with a range of straightforward combinations between FL methods and RL. The results demonstrate that SCCD has a superior performance in four classical continuous control tasks with non-IID environments." @default.
- W4381164013 created "2023-06-20" @default.
- W4381164013 creator A5003053702 @default.
- W4381164013 creator A5030222911 @default.
- W4381164013 creator A5034417471 @default.
- W4381164013 creator A5038516431 @default.
- W4381164013 creator A5047240103 @default.
- W4381164013 creator A5086330863 @default.
- W4381164013 date "2023-08-10" @default.
- W4381164013 modified "2023-09-27" @default.
- W4381164013 title "Server-Client Collaborative Distillation for Federated Reinforcement Learning" @default.
- W4381164013 cites W1641379095 @default.
- W4381164013 cites W1980516134 @default.
- W4381164013 cites W2137983211 @default.
- W4381164013 cites W2147492008 @default.
- W4381164013 cites W2165698076 @default.
- W4381164013 cites W2802595847 @default.
- W4381164013 cites W2968937098 @default.
- W4381164013 cites W3004397388 @default.
- W4381164013 cites W3025869479 @default.
- W4381164013 cites W3172419167 @default.
- W4381164013 cites W3182158470 @default.
- W4381164013 cites W3207250575 @default.
- W4381164013 cites W4220842548 @default.
- W4381164013 cites W4283790903 @default.
- W4381164013 cites W4284974283 @default.
- W4381164013 cites W4317815302 @default.
- W4381164013 cites W4362653493 @default.
- W4381164013 doi "https://doi.org/10.1145/3604939" @default.
- W4381164013 hasPublicationYear "2023" @default.
- W4381164013 type Work @default.
- W4381164013 citedByCount "0" @default.
- W4381164013 crossrefType "journal-article" @default.
- W4381164013 hasAuthorship W4381164013A5003053702 @default.
- W4381164013 hasAuthorship W4381164013A5030222911 @default.
- W4381164013 hasAuthorship W4381164013A5034417471 @default.
- W4381164013 hasAuthorship W4381164013A5038516431 @default.
- W4381164013 hasAuthorship W4381164013A5047240103 @default.
- W4381164013 hasAuthorship W4381164013A5086330863 @default.
- W4381164013 hasBestOaLocation W43811640131 @default.
- W4381164013 hasConcept C119857082 @default.
- W4381164013 hasConcept C121955636 @default.
- W4381164013 hasConcept C144133560 @default.
- W4381164013 hasConcept C154945302 @default.
- W4381164013 hasConcept C178790620 @default.
- W4381164013 hasConcept C185592680 @default.
- W4381164013 hasConcept C196083921 @default.
- W4381164013 hasConcept C202444582 @default.
- W4381164013 hasConcept C204030448 @default.
- W4381164013 hasConcept C2992525071 @default.
- W4381164013 hasConcept C33923547 @default.
- W4381164013 hasConcept C41008148 @default.
- W4381164013 hasConcept C9652623 @default.
- W4381164013 hasConcept C97541855 @default.
- W4381164013 hasConceptScore W4381164013C119857082 @default.
- W4381164013 hasConceptScore W4381164013C121955636 @default.
- W4381164013 hasConceptScore W4381164013C144133560 @default.
- W4381164013 hasConceptScore W4381164013C154945302 @default.
- W4381164013 hasConceptScore W4381164013C178790620 @default.
- W4381164013 hasConceptScore W4381164013C185592680 @default.
- W4381164013 hasConceptScore W4381164013C196083921 @default.
- W4381164013 hasConceptScore W4381164013C202444582 @default.
- W4381164013 hasConceptScore W4381164013C204030448 @default.
- W4381164013 hasConceptScore W4381164013C2992525071 @default.
- W4381164013 hasConceptScore W4381164013C33923547 @default.
- W4381164013 hasConceptScore W4381164013C41008148 @default.
- W4381164013 hasConceptScore W4381164013C9652623 @default.
- W4381164013 hasConceptScore W4381164013C97541855 @default.
- W4381164013 hasFunder F4320320955 @default.
- W4381164013 hasFunder F4320321001 @default.
- W4381164013 hasFunder F4320321885 @default.
- W4381164013 hasFunder F4320335777 @default.
- W4381164013 hasFunder F4320335787 @default.
- W4381164013 hasFunder F4320337111 @default.
- W4381164013 hasIssue "1" @default.
- W4381164013 hasLocation W43811640131 @default.
- W4381164013 hasOpenAccess W4381164013 @default.
- W4381164013 hasPrimaryLocation W43811640131 @default.
- W4381164013 hasRelatedWork W260766989 @default.
- W4381164013 hasRelatedWork W2959276766 @default.
- W4381164013 hasRelatedWork W2961085424 @default.
- W4381164013 hasRelatedWork W3074294383 @default.
- W4381164013 hasRelatedWork W3105298093 @default.
- W4381164013 hasRelatedWork W4206669594 @default.
- W4381164013 hasRelatedWork W4288828925 @default.
- W4381164013 hasRelatedWork W4295941380 @default.
- W4381164013 hasRelatedWork W4315629996 @default.
- W4381164013 hasRelatedWork W4319083788 @default.
- W4381164013 hasVolume "18" @default.
- W4381164013 isParatext "false" @default.
- W4381164013 isRetracted "false" @default.
- W4381164013 workType "article" @default.