Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384705413> ?p ?o ?g. }
Showing items 1 to 70 of
70
with 100 items per page.
- W4384705413 abstract "Federated Learning (FL) emerges as a distributed machine learning paradigm without end-user data transmission, effectively avoiding privacy leakage. Participating devices in FL are usually bandwidth-constrained, and the uplink is much slower than the downlink in wireless networks, which causes a severe uplink communication bottleneck. A prominent direction to alleviate this problem is federated dropout, which drops fractional weights of local models. However, existing federated dropout studies focus on random or ordered dropout and lack theoretical support, resulting in unguaranteed performance. In this paper, we propose Federated learning with Bayesian Inference-based Adaptive Dropout (FedBIAD), which regards weight rows of local models as probability distributions and adaptively drops partial weight rows based on importance indicators correlated with the trend of local training loss. By applying FedBIAD, each client adaptively selects a high-quality dropping pattern with accurate approximations and only transmits parameters of non-dropped weight rows to mitigate uplink costs while improving accuracy. Theoretical analysis demonstrates that the convergence rate of the average generalization error of FedBIAD is minimax optimal up to a squared logarithmic factor. Extensive experiments on image classification and next-word prediction show that compared with status quo approaches, FedBIAD provides 2× uplink reduction with an accuracy increase of up to 2.41% even on non-Independent and Identically Distributed (non-IID) data, which brings up to 72% decrease in training time." @default.
- W4384705413 created "2023-07-20" @default.
- W4384705413 creator A5008456610 @default.
- W4384705413 creator A5017353282 @default.
- W4384705413 creator A5033546667 @default.
- W4384705413 creator A5052821016 @default.
- W4384705413 creator A5080344490 @default.
- W4384705413 creator A5081973140 @default.
- W4384705413 date "2023-05-01" @default.
- W4384705413 modified "2023-09-24" @default.
- W4384705413 title "FedBIAD: Communication-Efficient and Accuracy-Guaranteed Federated Learning with Bayesian Inference-Based Adaptive Dropout" @default.
- W4384705413 cites W2982464076 @default.
- W4384705413 cites W2989289980 @default.
- W4384705413 cites W3045317090 @default.
- W4384705413 cites W3101177651 @default.
- W4384705413 cites W3184192107 @default.
- W4384705413 cites W3202260564 @default.
- W4384705413 cites W4283809296 @default.
- W4384705413 cites W4285504041 @default.
- W4384705413 cites W4289534065 @default.
- W4384705413 cites W4289546285 @default.
- W4384705413 doi "https://doi.org/10.1109/ipdps54959.2023.00056" @default.
- W4384705413 hasPublicationYear "2023" @default.
- W4384705413 type Work @default.
- W4384705413 citedByCount "0" @default.
- W4384705413 crossrefType "proceedings-article" @default.
- W4384705413 hasAuthorship W4384705413A5008456610 @default.
- W4384705413 hasAuthorship W4384705413A5017353282 @default.
- W4384705413 hasAuthorship W4384705413A5033546667 @default.
- W4384705413 hasAuthorship W4384705413A5052821016 @default.
- W4384705413 hasAuthorship W4384705413A5080344490 @default.
- W4384705413 hasAuthorship W4384705413A5081973140 @default.
- W4384705413 hasConcept C107673813 @default.
- W4384705413 hasConcept C11413529 @default.
- W4384705413 hasConcept C119857082 @default.
- W4384705413 hasConcept C138660444 @default.
- W4384705413 hasConcept C154945302 @default.
- W4384705413 hasConcept C160234255 @default.
- W4384705413 hasConcept C2776145597 @default.
- W4384705413 hasConcept C2776214188 @default.
- W4384705413 hasConcept C31258907 @default.
- W4384705413 hasConcept C41008148 @default.
- W4384705413 hasConceptScore W4384705413C107673813 @default.
- W4384705413 hasConceptScore W4384705413C11413529 @default.
- W4384705413 hasConceptScore W4384705413C119857082 @default.
- W4384705413 hasConceptScore W4384705413C138660444 @default.
- W4384705413 hasConceptScore W4384705413C154945302 @default.
- W4384705413 hasConceptScore W4384705413C160234255 @default.
- W4384705413 hasConceptScore W4384705413C2776145597 @default.
- W4384705413 hasConceptScore W4384705413C2776214188 @default.
- W4384705413 hasConceptScore W4384705413C31258907 @default.
- W4384705413 hasConceptScore W4384705413C41008148 @default.
- W4384705413 hasFunder F4320321001 @default.
- W4384705413 hasFunder F4320335777 @default.
- W4384705413 hasLocation W43847054131 @default.
- W4384705413 hasOpenAccess W4384705413 @default.
- W4384705413 hasPrimaryLocation W43847054131 @default.
- W4384705413 hasRelatedWork W2753218748 @default.
- W4384705413 hasRelatedWork W2774409638 @default.
- W4384705413 hasRelatedWork W2889562828 @default.
- W4384705413 hasRelatedWork W2952635455 @default.
- W4384705413 hasRelatedWork W2952801366 @default.
- W4384705413 hasRelatedWork W2953280030 @default.
- W4384705413 hasRelatedWork W3002319139 @default.
- W4384705413 hasRelatedWork W3125702521 @default.
- W4384705413 hasRelatedWork W3141804903 @default.
- W4384705413 hasRelatedWork W4292442683 @default.
- W4384705413 isParatext "false" @default.
- W4384705413 isRetracted "false" @default.
- W4384705413 workType "article" @default.