Matches in SemOpenAlex for { <https://semopenalex.org/work/W3199025691> ?p ?o ?g. }
- W3199025691 abstract "Federated Learning (FL) has shown great potential as a privacy-preserving solution to learning from decentralized data that are only accessible to end devices (i.e., clients). In many scenarios, however, a large proportion of the clients are probably in possession of low-quality data that are biased, noisy or even irrelevant. As a result, they could significantly slow down the convergence of the global model we aim to build and also compromise its quality. In light of this, we propose FedProf, a novel algorithm for optimizing FL under such circumstances without breaching data privacy. The key of our approach is a distributional representation profiling and matching scheme that uses the global model to dynamically profile data representations and allows for low-cost, lightweight representation matching. Based on the scheme we adaptively score each client and adjust its participation probability so as to mitigate the impact of low-value clients on the training process. We have conducted extensive experiments on public datasets using various FL settings. The results show that the selective behaviour of our algorithm leads to a significant reduction in the number of communication rounds and the amount of time (up to 2.4x speedup) for the global model to converge and also provides accuracy gain." @default.
- W3199025691 created "2021-09-27" @default.
- W3199025691 creator A5007440245 @default.
- W3199025691 creator A5030234311 @default.
- W3199025691 creator A5047828703 @default.
- W3199025691 creator A5064596191 @default.
- W3199025691 date "2021-02-02" @default.
- W3199025691 modified "2023-10-18" @default.
- W3199025691 title "FedProf: Selective Federated Learning with Representation Profiling" @default.
- W3199025691 cites W1768994003 @default.
- W3199025691 cites W1994786006 @default.
- W3199025691 cites W2031533839 @default.
- W3199025691 cites W204053018 @default.
- W3199025691 cites W2043623344 @default.
- W3199025691 cites W2163922914 @default.
- W3199025691 cites W2168290280 @default.
- W3199025691 cites W2194775991 @default.
- W3199025691 cites W2535838896 @default.
- W3199025691 cites W2541884796 @default.
- W3199025691 cites W2612634114 @default.
- W3199025691 cites W2626580042 @default.
- W3199025691 cites W2769644379 @default.
- W3199025691 cites W2890924858 @default.
- W3199025691 cites W2896422817 @default.
- W3199025691 cites W2900120080 @default.
- W3199025691 cites W2920095265 @default.
- W3199025691 cites W2942571707 @default.
- W3199025691 cites W2942810103 @default.
- W3199025691 cites W2962688627 @default.
- W3199025691 cites W2963318081 @default.
- W3199025691 cites W2963422767 @default.
- W3199025691 cites W2963691377 @default.
- W3199025691 cites W2963759780 @default.
- W3199025691 cites W2964163156 @default.
- W3199025691 cites W2975985894 @default.
- W3199025691 cites W2982464076 @default.
- W3199025691 cites W2995022099 @default.
- W3199025691 cites W2995653155 @default.
- W3199025691 cites W3004117333 @default.
- W3199025691 cites W3034938110 @default.
- W3199025691 cites W3036468020 @default.
- W3199025691 cites W3037024761 @default.
- W3199025691 cites W3037871107 @default.
- W3199025691 cites W3043758338 @default.
- W3199025691 cites W3045611708 @default.
- W3199025691 cites W3046056026 @default.
- W3199025691 cites W3047304572 @default.
- W3199025691 cites W3048715803 @default.
- W3199025691 cites W3089426191 @default.
- W3199025691 cites W3099314130 @default.
- W3199025691 cites W3102074914 @default.
- W3199025691 cites W3105122387 @default.
- W3199025691 cites W3106416029 @default.
- W3199025691 cites W3111182214 @default.
- W3199025691 cites W3126083338 @default.
- W3199025691 cites W3127965231 @default.
- W3199025691 cites W3156841666 @default.
- W3199025691 cites W3172811949 @default.
- W3199025691 cites W3174188899 @default.
- W3199025691 cites W3182158470 @default.
- W3199025691 cites W4221030344 @default.
- W3199025691 doi "https://doi.org/10.48550/arxiv.2102.01733" @default.
- W3199025691 hasPublicationYear "2021" @default.
- W3199025691 type Work @default.
- W3199025691 sameAs 3199025691 @default.
- W3199025691 citedByCount "0" @default.
- W3199025691 crossrefType "posted-content" @default.
- W3199025691 hasAuthorship W3199025691A5007440245 @default.
- W3199025691 hasAuthorship W3199025691A5030234311 @default.
- W3199025691 hasAuthorship W3199025691A5047828703 @default.
- W3199025691 hasAuthorship W3199025691A5064596191 @default.
- W3199025691 hasBestOaLocation W31990256911 @default.
- W3199025691 hasConcept C105795698 @default.
- W3199025691 hasConcept C111919701 @default.
- W3199025691 hasConcept C119857082 @default.
- W3199025691 hasConcept C124101348 @default.
- W3199025691 hasConcept C144024400 @default.
- W3199025691 hasConcept C154945302 @default.
- W3199025691 hasConcept C165064840 @default.
- W3199025691 hasConcept C17744445 @default.
- W3199025691 hasConcept C187191949 @default.
- W3199025691 hasConcept C199539241 @default.
- W3199025691 hasConcept C2776359362 @default.
- W3199025691 hasConcept C2992525071 @default.
- W3199025691 hasConcept C33923547 @default.
- W3199025691 hasConcept C36289849 @default.
- W3199025691 hasConcept C41008148 @default.
- W3199025691 hasConcept C46355384 @default.
- W3199025691 hasConcept C68339613 @default.
- W3199025691 hasConcept C94625758 @default.
- W3199025691 hasConceptScore W3199025691C105795698 @default.
- W3199025691 hasConceptScore W3199025691C111919701 @default.
- W3199025691 hasConceptScore W3199025691C119857082 @default.
- W3199025691 hasConceptScore W3199025691C124101348 @default.
- W3199025691 hasConceptScore W3199025691C144024400 @default.
- W3199025691 hasConceptScore W3199025691C154945302 @default.
- W3199025691 hasConceptScore W3199025691C165064840 @default.
- W3199025691 hasConceptScore W3199025691C17744445 @default.
- W3199025691 hasConceptScore W3199025691C187191949 @default.
- W3199025691 hasConceptScore W3199025691C199539241 @default.