Matches in SemOpenAlex for { <https://semopenalex.org/work/W3090223552> ?p ?o ?g. }
- W3090223552 abstract "Secure model aggregation across many users is a key component of federated learning systems. The state-of-the-art protocols for secure model aggregation, which are based on additive masking, require all users to quantize their model updates to the same level of quantization. This severely degrades their performance due to lack of adaptation to available bandwidth at different users. We propose three schemes that allow secure model aggregation while using heterogeneous quantization. This enables the users to adjust their quantization proportional to their available bandwidth, which can provide a substantially better trade-off between the accuracy of training and the communication time. The proposed schemes are based on a grouping strategy by partitioning the network into groups, and partitioning the local model updates of users into segments. Instead of applying aggregation protocol to the entire local model update vector, it is applied on segments with specific coordination between users. We theoretically evaluate the quantization error for our schemes, and also demonstrate how our schemes can be utilized to overcome Byzantine users." @default.
- W3090223552 created "2020-10-08" @default.
- W3090223552 creator A5047191296 @default.
- W3090223552 creator A5083855675 @default.
- W3090223552 date "2020-09-30" @default.
- W3090223552 modified "2023-09-27" @default.
- W3090223552 title "Secure Aggregation with Heterogeneous Quantization in Federated Learning." @default.
- W3090223552 cites W1519947574 @default.
- W3090223552 cites W2006453614 @default.
- W3090223552 cites W2027471022 @default.
- W3090223552 cites W2092422002 @default.
- W3090223552 cites W2141420453 @default.
- W3090223552 cites W2156186849 @default.
- W3090223552 cites W2405578611 @default.
- W3090223552 cites W2407022425 @default.
- W3090223552 cites W2535690855 @default.
- W3090223552 cites W2535838896 @default.
- W3090223552 cites W2541884796 @default.
- W3090223552 cites W2547352193 @default.
- W3090223552 cites W2752689052 @default.
- W3090223552 cites W2766140019 @default.
- W3090223552 cites W2767079719 @default.
- W3090223552 cites W2769644379 @default.
- W3090223552 cites W2799803467 @default.
- W3090223552 cites W2946633206 @default.
- W3090223552 cites W2948579237 @default.
- W3090223552 cites W2963766684 @default.
- W3090223552 cites W2964004663 @default.
- W3090223552 cites W2964279249 @default.
- W3090223552 cites W2970408908 @default.
- W3090223552 cites W2972570881 @default.
- W3090223552 cites W2989289980 @default.
- W3090223552 cites W2995022099 @default.
- W3090223552 cites W3012798438 @default.
- W3090223552 cites W3013310637 @default.
- W3090223552 cites W3013730797 @default.
- W3090223552 cites W3015274184 @default.
- W3090223552 cites W3017123256 @default.
- W3090223552 cites W3035498193 @default.
- W3090223552 cites W3038028469 @default.
- W3090223552 cites W3088123628 @default.
- W3090223552 cites W3096328345 @default.
- W3090223552 cites W3105285631 @default.
- W3090223552 cites W3107632872 @default.
- W3090223552 cites W3113297383 @default.
- W3090223552 cites W3124562976 @default.
- W3090223552 cites W3137092842 @default.
- W3090223552 cites W3175192640 @default.
- W3090223552 hasPublicationYear "2020" @default.
- W3090223552 type Work @default.
- W3090223552 sameAs 3090223552 @default.
- W3090223552 citedByCount "8" @default.
- W3090223552 countsByYear W30902235522020 @default.
- W3090223552 countsByYear W30902235522021 @default.
- W3090223552 countsByYear W30902235522022 @default.
- W3090223552 crossrefType "posted-content" @default.
- W3090223552 hasAuthorship W3090223552A5047191296 @default.
- W3090223552 hasAuthorship W3090223552A5083855675 @default.
- W3090223552 hasConcept C11413529 @default.
- W3090223552 hasConcept C120314980 @default.
- W3090223552 hasConcept C154945302 @default.
- W3090223552 hasConcept C199833920 @default.
- W3090223552 hasConcept C28855332 @default.
- W3090223552 hasConcept C2992525071 @default.
- W3090223552 hasConcept C31258907 @default.
- W3090223552 hasConcept C41008148 @default.
- W3090223552 hasConcept C80444323 @default.
- W3090223552 hasConceptScore W3090223552C11413529 @default.
- W3090223552 hasConceptScore W3090223552C120314980 @default.
- W3090223552 hasConceptScore W3090223552C154945302 @default.
- W3090223552 hasConceptScore W3090223552C199833920 @default.
- W3090223552 hasConceptScore W3090223552C28855332 @default.
- W3090223552 hasConceptScore W3090223552C2992525071 @default.
- W3090223552 hasConceptScore W3090223552C31258907 @default.
- W3090223552 hasConceptScore W3090223552C41008148 @default.
- W3090223552 hasConceptScore W3090223552C80444323 @default.
- W3090223552 hasLocation W30902235521 @default.
- W3090223552 hasOpenAccess W3090223552 @default.
- W3090223552 hasPrimaryLocation W30902235521 @default.
- W3090223552 hasRelatedWork W2024710115 @default.
- W3090223552 hasRelatedWork W2511985007 @default.
- W3090223552 hasRelatedWork W2541884796 @default.
- W3090223552 hasRelatedWork W2767079719 @default.
- W3090223552 hasRelatedWork W2767593597 @default.
- W3090223552 hasRelatedWork W2900319533 @default.
- W3090223552 hasRelatedWork W2902113386 @default.
- W3090223552 hasRelatedWork W2989253922 @default.
- W3090223552 hasRelatedWork W2995022099 @default.
- W3090223552 hasRelatedWork W3008187686 @default.
- W3090223552 hasRelatedWork W3014367186 @default.
- W3090223552 hasRelatedWork W3044211235 @default.
- W3090223552 hasRelatedWork W3103802018 @default.
- W3090223552 hasRelatedWork W3107632872 @default.
- W3090223552 hasRelatedWork W3114967426 @default.
- W3090223552 hasRelatedWork W3181552337 @default.
- W3090223552 hasRelatedWork W3182946142 @default.
- W3090223552 hasRelatedWork W3183596499 @default.
- W3090223552 hasRelatedWork W3189092282 @default.
- W3090223552 hasRelatedWork W2555962320 @default.
- W3090223552 isParatext "false" @default.