File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1109/IWCMC58020.2023.10183093
- Scopus: eid_2-s2.0-85167737711
Supplementary
-
Citations:
- Scopus: 0
- Appears in Collections:
Conference Paper: Convergence Analysis for Wireless Federated Learning with Gradient Recycling
Title | Convergence Analysis for Wireless Federated Learning with Gradient Recycling |
---|---|
Authors | |
Keywords | Federated Learning unreliable transmission |
Issue Date | 2023 |
Citation | 2023 International Wireless Communications and Mobile Computing, IWCMC 2023, 2023, p. 1232-1237 How to Cite? |
Abstract | How to tackle the unreliability in wireless channels is critical for federated learning (FL). To solve this problem, we propose a novel FL framework, namely FL with gradient recycling (FL-GR), which recycles the historical gradients of unscheduled and transmission-failure devices to improve the learning performance of FL. Based on the proposed FL-GR, we theoretically analyze how the wireless network parameters affect the convergence bound of FL-GR, revealing that scheduling devices with large staleness and increasing their transmit power in each round helps improve learning performance. Simulation results on MNIST and CIFAR-10 show that FL-GR is able to achieve higher accuracy and fast convergence speed than conventional FL algorithms without gradient recycling. |
Persistent Identifier | http://hdl.handle.net/10722/349946 |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Chen, Zhixiong | - |
dc.contributor.author | Yi, Wenqiang | - |
dc.contributor.author | Liu, Yuanwei | - |
dc.contributor.author | Nallanathan, Arumugam | - |
dc.date.accessioned | 2024-10-17T07:02:03Z | - |
dc.date.available | 2024-10-17T07:02:03Z | - |
dc.date.issued | 2023 | - |
dc.identifier.citation | 2023 International Wireless Communications and Mobile Computing, IWCMC 2023, 2023, p. 1232-1237 | - |
dc.identifier.uri | http://hdl.handle.net/10722/349946 | - |
dc.description.abstract | How to tackle the unreliability in wireless channels is critical for federated learning (FL). To solve this problem, we propose a novel FL framework, namely FL with gradient recycling (FL-GR), which recycles the historical gradients of unscheduled and transmission-failure devices to improve the learning performance of FL. Based on the proposed FL-GR, we theoretically analyze how the wireless network parameters affect the convergence bound of FL-GR, revealing that scheduling devices with large staleness and increasing their transmit power in each round helps improve learning performance. Simulation results on MNIST and CIFAR-10 show that FL-GR is able to achieve higher accuracy and fast convergence speed than conventional FL algorithms without gradient recycling. | - |
dc.language | eng | - |
dc.relation.ispartof | 2023 International Wireless Communications and Mobile Computing, IWCMC 2023 | - |
dc.subject | Federated Learning | - |
dc.subject | unreliable transmission | - |
dc.title | Convergence Analysis for Wireless Federated Learning with Gradient Recycling | - |
dc.type | Conference_Paper | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1109/IWCMC58020.2023.10183093 | - |
dc.identifier.scopus | eid_2-s2.0-85167737711 | - |
dc.identifier.spage | 1232 | - |
dc.identifier.epage | 1237 | - |