Matches in SemOpenAlex for { <https://semopenalex.org/work/W3121861147> ?p ?o ?g. }
Showing items 1 to 65 of
65
with 100 items per page.
- W3121861147 abstract "Federated learning (FL) is a framework for distributed learning of centralized models. In FL, a set of edge devices train a model using their local data, while repeatedly exchanging their trained model with a central server, allowing to tune a global model without having the users share their possibly private data. A major challenge in FL is to reduce the bandwidth and energy consumption due to the repeated transmissions of large volumes of data by a large number of users over the wireless channel. Recently, over-the-air (OTA) FL has been suggested to achieve this goal. In this setting, all users transmit their data signal simultaneously over a Multiple Access Channel (MAC), and the computation is done over the wireless channel. In this paper, we develop a novel convergent OTA FL (COTAF) algorithm, which induces precoding and scaling upon transmissions to gradually mitigate the effect of the noisy channel, thus facilitating FL convergence. We analyze the convergence of COTAF to the loss minimizing model theoretically, showing its ability to achieve a convergence rate similar to that achievable over error-free channels. Our simulations demonstrate the improved convergence of COTAF for training using non-synthetic datasets." @default.
- W3121861147 created "2021-02-01" @default.
- W3121861147 creator A5005913897 @default.
- W3121861147 creator A5012405763 @default.
- W3121861147 creator A5058072770 @default.
- W3121861147 creator A5066085947 @default.
- W3121861147 date "2020-12-01" @default.
- W3121861147 modified "2023-10-17" @default.
- W3121861147 title "COTAF: Convergent Over-the-Air Federated Learning" @default.
- W3121861147 cites W1989888201 @default.
- W3121861147 cites W1991349164 @default.
- W3121861147 cites W2027531092 @default.
- W3121861147 cites W2039503172 @default.
- W3121861147 cites W2140473338 @default.
- W3121861147 cites W2141002933 @default.
- W3121861147 cites W2153757022 @default.
- W3121861147 cites W2170597157 @default.
- W3121861147 cites W2463103808 @default.
- W3121861147 cites W2907379776 @default.
- W3121861147 cites W2919115771 @default.
- W3121861147 cites W2950044985 @default.
- W3121861147 cites W2960833983 @default.
- W3121861147 cites W2969343398 @default.
- W3121861147 cites W3006919779 @default.
- W3121861147 cites W3015901293 @default.
- W3121861147 cites W3081178085 @default.
- W3121861147 cites W3101036738 @default.
- W3121861147 cites W3103802018 @default.
- W3121861147 cites W595252221 @default.
- W3121861147 doi "https://doi.org/10.1109/globecom42002.2020.9322580" @default.
- W3121861147 hasPublicationYear "2020" @default.
- W3121861147 type Work @default.
- W3121861147 sameAs 3121861147 @default.
- W3121861147 citedByCount "16" @default.
- W3121861147 countsByYear W31218611472020 @default.
- W3121861147 countsByYear W31218611472021 @default.
- W3121861147 countsByYear W31218611472022 @default.
- W3121861147 countsByYear W31218611472023 @default.
- W3121861147 crossrefType "proceedings-article" @default.
- W3121861147 hasAuthorship W3121861147A5005913897 @default.
- W3121861147 hasAuthorship W3121861147A5012405763 @default.
- W3121861147 hasAuthorship W3121861147A5058072770 @default.
- W3121861147 hasAuthorship W3121861147A5066085947 @default.
- W3121861147 hasConcept C41008148 @default.
- W3121861147 hasConceptScore W3121861147C41008148 @default.
- W3121861147 hasFunder F4320308642 @default.
- W3121861147 hasFunder F4320308960 @default.
- W3121861147 hasFunder F4320322252 @default.
- W3121861147 hasLocation W31218611471 @default.
- W3121861147 hasOpenAccess W3121861147 @default.
- W3121861147 hasPrimaryLocation W31218611471 @default.
- W3121861147 hasRelatedWork W2049775471 @default.
- W3121861147 hasRelatedWork W2093578348 @default.
- W3121861147 hasRelatedWork W2350741829 @default.
- W3121861147 hasRelatedWork W2358668433 @default.
- W3121861147 hasRelatedWork W2376932109 @default.
- W3121861147 hasRelatedWork W2382290278 @default.
- W3121861147 hasRelatedWork W2390279801 @default.
- W3121861147 hasRelatedWork W2748952813 @default.
- W3121861147 hasRelatedWork W2899084033 @default.
- W3121861147 hasRelatedWork W3004735627 @default.
- W3121861147 isParatext "false" @default.
- W3121861147 isRetracted "false" @default.
- W3121861147 magId "3121861147" @default.
- W3121861147 workType "article" @default.