DC Field | Value | Language |
---|---|---|
dc.contributor.author | YONGJEONG, OH | - |
dc.contributor.author | JEON, YO SEB | - |
dc.contributor.author | Chen, Mingzhe | - |
dc.contributor.author | Saad, Walid | - |
dc.date.accessioned | 2023-09-01T07:20:28Z | - |
dc.date.available | 2023-09-01T07:20:28Z | - |
dc.date.created | 2023-07-21 | - |
dc.date.issued | 2024-03 | - |
dc.identifier.issn | 1536-1276 | - |
dc.identifier.uri | https://oasis.postech.ac.kr/handle/2014.oak/118502 | - |
dc.description.abstract | In this paper, a new communication-efficient federated learning (FL) framework is proposed, inspired by vector quantized compressed sensing. The basic strategy of the proposed framework is to compress the local model update at each device by applying dimensionality reduction followed by vector quantization. Subsequently, the global model update is reconstructed at a parameter server by applying a sparse signal recovery algorithm to the aggregation of the compressed local model updates. By harnessing the benefits of both dimensionality reduction and vector quantization, the proposed framework effectively reduces the communication overhead of local update transmissions. Both the design of the vector quantizer and the key parameters for the compression are optimized so as to minimize the reconstruction error of the global model update under the constraint of wireless link capacity. By considering the reconstruction error, the convergence rate of the proposed framework is also analyzed for a non-convex loss function. Simulation results on the MNIST and FEMNIST datasets demonstrate that the proposed framework provides more than a 2.4% increase in classification accuracy compared to state-of-the-art FL frameworks when the communication overhead of the local model update transmission is 0.1 bit per local model entry. | - |
dc.language | English | - |
dc.publisher | Institute of Electrical and Electronics Engineers | - |
dc.relation.isPartOf | IEEE Transactions on Wireless Communications | - |
dc.title | FedVQCS: Federated Learning via Vector Quantized Compressed Sensing | - |
dc.type | Article | - |
dc.identifier.doi | 10.1109/twc.2023.3291877 | - |
dc.type.rims | ART | - |
dc.identifier.bibliographicCitation | IEEE Transactions on Wireless Communications, v.23, no.3, pp.1 - 1 | - |
dc.identifier.wosid | 001196528600048 | - |
dc.citation.endPage | 1 | - |
dc.citation.number | 3 | - |
dc.citation.startPage | 1 | - |
dc.citation.title | IEEE Transactions on Wireless Communications | - |
dc.citation.volume | 23 | - |
dc.contributor.affiliatedAuthor | YONGJEONG, OH | - |
dc.contributor.affiliatedAuthor | JEON, YO SEB | - |
dc.identifier.scopusid | 2-s2.0-85164683558 | - |
dc.description.journalClass | 1 | - |
dc.description.journalClass | 1 | - |
dc.description.isOpenAccess | N | - |
dc.type.docType | Article | - |
dc.subject.keywordAuthor | dimensionality reduction | - |
dc.subject.keywordAuthor | distributed learning | - |
dc.subject.keywordAuthor | Federated learning | - |
dc.subject.keywordAuthor | quantized compressed sensing | - |
dc.subject.keywordAuthor | vector quantization | - |
dc.relation.journalWebOfScienceCategory | Engineering, Electrical & Electronic | - |
dc.relation.journalWebOfScienceCategory | Telecommunications | - |
dc.description.journalRegisteredClass | scie | - |
dc.description.journalRegisteredClass | scopus | - |
dc.relation.journalResearchArea | Engineering | - |
dc.relation.journalResearchArea | Telecommunications | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
library@postech.ac.kr Tel: 054-279-2548
Copyrights © by 2017 Pohang University of Science ad Technology All right reserved.