Matches in SemOpenAlex for { <https://semopenalex.org/work/W2907379776> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W2907379776 endingPage "2169" @default.
- W2907379776 startingPage "2155" @default.
- W2907379776 abstract "We study federated machine learning (ML) at the wireless edge, where power- and bandwidth-limited wireless devices with local datasets carry out distributed stochastic gradient descent (DSGD) with the help of a remote parameter server (PS). Standard approaches assume separate computation and communication, where local gradient estimates are compressed and transmitted to the PS over orthogonal links. Following this digital approach, we introduce D-DSGD, in which the wireless devices employ gradient quantization and error accumulation, and transmit their gradient estimates to the PS over a multiple access channel (MAC). We then introduce a novel analog scheme, called A-DSGD, which exploits the additive nature of the wireless MAC for over-the-air gradient computation, and provide convergence analysis for this approach. In A-DSGD, the devices first sparsify their gradient estimates, and then project them to a lower dimensional space imposed by the available channel bandwidth. These projections are sent directly over the MAC without employing any digital code. Numerical results show that A-DSGD converges faster than D-DSGD thanks to its more efficient use of the limited bandwidth and the natural alignment of the gradient estimates over the channel. The improvement is particularly compelling at low power and low bandwidth regimes. We also illustrate for a classification problem that, A-DSGD is more robust to bias in data distribution across devices, while D-DSGD significantly outperforms other digital schemes in the literature. We also observe that both D-DSGD and A-DSGD perform better by increasing the number of devices (while keeping the total dataset size constant), showing their ability in harnessing the computation power of edge devices." @default.
- W2907379776 created "2019-01-11" @default.
- W2907379776 creator A5010987707 @default.
- W2907379776 creator A5016883501 @default.
- W2907379776 date "2020-01-01" @default.
- W2907379776 modified "2023-10-18" @default.
- W2907379776 title "Machine Learning at the Wireless Edge: Distributed Stochastic Gradient Descent Over-the-Air" @default.
- W2907379776 cites W1837352444 @default.
- W2907379776 cites W1981573335 @default.
- W2907379776 cites W2082029531 @default.
- W2907379776 cites W2110457001 @default.
- W2907379776 cites W2128011496 @default.
- W2907379776 cites W2552857149 @default.
- W2907379776 cites W2781017182 @default.
- W2907379776 cites W2898338514 @default.
- W2907379776 cites W2971330699 @default.
- W2907379776 cites W2975064142 @default.
- W2907379776 cites W2978015420 @default.
- W2907379776 cites W2981138228 @default.
- W2907379776 cites W2985108934 @default.
- W2907379776 cites W2991308345 @default.
- W2907379776 cites W2999074226 @default.
- W2907379776 cites W3003959415 @default.
- W2907379776 cites W3006919779 @default.
- W2907379776 cites W3098848552 @default.
- W2907379776 cites W3101036738 @default.
- W2907379776 cites W4231394494 @default.
- W2907379776 doi "https://doi.org/10.1109/tsp.2020.2981904" @default.
- W2907379776 hasPublicationYear "2020" @default.
- W2907379776 type Work @default.
- W2907379776 sameAs 2907379776 @default.
- W2907379776 citedByCount "350" @default.
- W2907379776 countsByYear W29073797762018 @default.
- W2907379776 countsByYear W29073797762019 @default.
- W2907379776 countsByYear W29073797762020 @default.
- W2907379776 countsByYear W29073797762021 @default.
- W2907379776 countsByYear W29073797762022 @default.
- W2907379776 countsByYear W29073797762023 @default.
- W2907379776 crossrefType "journal-article" @default.
- W2907379776 hasAuthorship W2907379776A5010987707 @default.
- W2907379776 hasAuthorship W2907379776A5016883501 @default.
- W2907379776 hasBestOaLocation W29073797762 @default.
- W2907379776 hasConcept C108037233 @default.
- W2907379776 hasConcept C11413529 @default.
- W2907379776 hasConcept C127162648 @default.
- W2907379776 hasConcept C153258448 @default.
- W2907379776 hasConcept C154945302 @default.
- W2907379776 hasConcept C206688291 @default.
- W2907379776 hasConcept C2776257435 @default.
- W2907379776 hasConcept C28855332 @default.
- W2907379776 hasConcept C31258907 @default.
- W2907379776 hasConcept C41008148 @default.
- W2907379776 hasConcept C45374587 @default.
- W2907379776 hasConcept C50644808 @default.
- W2907379776 hasConcept C555944384 @default.
- W2907379776 hasConcept C76155785 @default.
- W2907379776 hasConceptScore W2907379776C108037233 @default.
- W2907379776 hasConceptScore W2907379776C11413529 @default.
- W2907379776 hasConceptScore W2907379776C127162648 @default.
- W2907379776 hasConceptScore W2907379776C153258448 @default.
- W2907379776 hasConceptScore W2907379776C154945302 @default.
- W2907379776 hasConceptScore W2907379776C206688291 @default.
- W2907379776 hasConceptScore W2907379776C2776257435 @default.
- W2907379776 hasConceptScore W2907379776C28855332 @default.
- W2907379776 hasConceptScore W2907379776C31258907 @default.
- W2907379776 hasConceptScore W2907379776C41008148 @default.
- W2907379776 hasConceptScore W2907379776C45374587 @default.
- W2907379776 hasConceptScore W2907379776C50644808 @default.
- W2907379776 hasConceptScore W2907379776C555944384 @default.
- W2907379776 hasConceptScore W2907379776C76155785 @default.
- W2907379776 hasFunder F4320334678 @default.
- W2907379776 hasLocation W29073797761 @default.
- W2907379776 hasLocation W29073797762 @default.
- W2907379776 hasLocation W29073797763 @default.
- W2907379776 hasOpenAccess W2907379776 @default.
- W2907379776 hasPrimaryLocation W29073797761 @default.
- W2907379776 hasRelatedWork W2248393001 @default.
- W2907379776 hasRelatedWork W2379719268 @default.
- W2907379776 hasRelatedWork W2905162178 @default.
- W2907379776 hasRelatedWork W2964163156 @default.
- W2907379776 hasRelatedWork W3092345941 @default.
- W2907379776 hasRelatedWork W3129457029 @default.
- W2907379776 hasRelatedWork W3205806653 @default.
- W2907379776 hasRelatedWork W3206635749 @default.
- W2907379776 hasRelatedWork W4289123174 @default.
- W2907379776 hasRelatedWork W1968689511 @default.
- W2907379776 hasVolume "68" @default.
- W2907379776 isParatext "false" @default.
- W2907379776 isRetracted "false" @default.
- W2907379776 magId "2907379776" @default.
- W2907379776 workType "article" @default.