Matches in SemOpenAlex for { <https://semopenalex.org/work/W3208283650> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W3208283650 endingPage "9054" @default.
- W3208283650 startingPage "9046" @default.
- W3208283650 abstract "The paradigm of Federated learning (FL) deals with multiple clients participating in collaborative training of a machine learning model under the orchestration of a central server. In this setup, each client’s data is private to itself and is not transferable to other clients or the server. Though FL paradigm has received significant interest recently from the research community, the problem of selecting the relevant clients w.r.t. the central server's learning objective is under-explored. We refer to these problems as Federated Relevant Client Selection (FRCS). Because the server doesn't have explicit control over the nature of data possessed by each client, the problem of selecting relevant clients is significantly complex in FL settings. In this paper, we resolve important and related FRCS problems viz., selecting clients with relevant data, detecting clients that possess data relevant to a particular target label, and rectifying corrupted data samples of individual clients. We follow a principled approach to address the above FRCS problems and develop a new federated learning method using the Shapley value concept from cooperative game theory. Towards this end, we propose a cooperative game involving the gradients shared by the clients. Using this game, we compute Shapley values of clients and then present Shapley value based Federated Averaging (S-FedAvg) algorithm that empowers the server to select relevant clients with high probability. S-FedAvg turns out to be critical in designing specific algorithms to address the FRCS problems. We finally conduct a thorough empirical analysis on image classification and speech recognition tasks to show the superior performance of S-FedAvg than the baselines in the context of supervised federated learning settings." @default.
- W3208283650 created "2021-11-08" @default.
- W3208283650 creator A5024416733 @default.
- W3208283650 creator A5071999741 @default.
- W3208283650 date "2021-05-18" @default.
- W3208283650 modified "2023-09-29" @default.
- W3208283650 title "Game of Gradients: Mitigating Irrelevant Clients in Federated Learning" @default.
- W3208283650 cites W2085648025 @default.
- W3208283650 cites W2110667918 @default.
- W3208283650 cites W2163300279 @default.
- W3208283650 cites W2473418344 @default.
- W3208283650 cites W2541884796 @default.
- W3208283650 cites W2752689052 @default.
- W3208283650 cites W2777914285 @default.
- W3208283650 cites W2807006176 @default.
- W3208283650 cites W2912213068 @default.
- W3208283650 cites W2913570153 @default.
- W3208283650 cites W2949574275 @default.
- W3208283650 cites W2951781666 @default.
- W3208283650 cites W2954034773 @default.
- W3208283650 cites W2955213239 @default.
- W3208283650 cites W2962741697 @default.
- W3208283650 cites W2964155802 @default.
- W3208283650 cites W2964292098 @default.
- W3208283650 cites W2970408908 @default.
- W3208283650 cites W2971221070 @default.
- W3208283650 cites W2978422189 @default.
- W3208283650 cites W2982654255 @default.
- W3208283650 cites W3001989995 @default.
- W3208283650 cites W3008477738 @default.
- W3208283650 cites W3034868650 @default.
- W3208283650 cites W3046219242 @default.
- W3208283650 cites W3106539628 @default.
- W3208283650 doi "https://doi.org/10.1609/aaai.v35i10.17093" @default.
- W3208283650 hasPublicationYear "2021" @default.
- W3208283650 type Work @default.
- W3208283650 sameAs 3208283650 @default.
- W3208283650 citedByCount "19" @default.
- W3208283650 countsByYear W32082836502022 @default.
- W3208283650 countsByYear W32082836502023 @default.
- W3208283650 crossrefType "journal-article" @default.
- W3208283650 hasAuthorship W3208283650A5024416733 @default.
- W3208283650 hasAuthorship W3208283650A5071999741 @default.
- W3208283650 hasBestOaLocation W32082836501 @default.
- W3208283650 hasConcept C119857082 @default.
- W3208283650 hasConcept C142362112 @default.
- W3208283650 hasConcept C153349607 @default.
- W3208283650 hasConcept C154945302 @default.
- W3208283650 hasConcept C162324750 @default.
- W3208283650 hasConcept C175444787 @default.
- W3208283650 hasConcept C177142836 @default.
- W3208283650 hasConcept C199022921 @default.
- W3208283650 hasConcept C199168358 @default.
- W3208283650 hasConcept C2775924081 @default.
- W3208283650 hasConcept C2776291640 @default.
- W3208283650 hasConcept C41008148 @default.
- W3208283650 hasConcept C558565934 @default.
- W3208283650 hasConceptScore W3208283650C119857082 @default.
- W3208283650 hasConceptScore W3208283650C142362112 @default.
- W3208283650 hasConceptScore W3208283650C153349607 @default.
- W3208283650 hasConceptScore W3208283650C154945302 @default.
- W3208283650 hasConceptScore W3208283650C162324750 @default.
- W3208283650 hasConceptScore W3208283650C175444787 @default.
- W3208283650 hasConceptScore W3208283650C177142836 @default.
- W3208283650 hasConceptScore W3208283650C199022921 @default.
- W3208283650 hasConceptScore W3208283650C199168358 @default.
- W3208283650 hasConceptScore W3208283650C2775924081 @default.
- W3208283650 hasConceptScore W3208283650C2776291640 @default.
- W3208283650 hasConceptScore W3208283650C41008148 @default.
- W3208283650 hasConceptScore W3208283650C558565934 @default.
- W3208283650 hasIssue "10" @default.
- W3208283650 hasLocation W32082836501 @default.
- W3208283650 hasLocation W32082836502 @default.
- W3208283650 hasOpenAccess W3208283650 @default.
- W3208283650 hasPrimaryLocation W32082836501 @default.
- W3208283650 hasRelatedWork W2081035100 @default.
- W3208283650 hasRelatedWork W2768920865 @default.
- W3208283650 hasRelatedWork W2961085424 @default.
- W3208283650 hasRelatedWork W3037819952 @default.
- W3208283650 hasRelatedWork W3187814221 @default.
- W3208283650 hasRelatedWork W4285260836 @default.
- W3208283650 hasRelatedWork W4286629047 @default.
- W3208283650 hasRelatedWork W4306321456 @default.
- W3208283650 hasRelatedWork W4306674287 @default.
- W3208283650 hasRelatedWork W4224009465 @default.
- W3208283650 hasVolume "35" @default.
- W3208283650 isParatext "false" @default.
- W3208283650 isRetracted "false" @default.
- W3208283650 magId "3208283650" @default.
- W3208283650 workType "article" @default.