基于梯度稀疏的盲联合边缘学习最优MIMO组合

2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC) Pub Date : 2022-03-24 DOI:10.48550/arXiv.2203.12957

Ema Becirovic, Zheng Chen, E. Larsson

{"title":"基于梯度稀疏的盲联合边缘学习最优MIMO组合","authors":"Ema Becirovic, Zheng Chen, E. Larsson","doi":"10.48550/arXiv.2203.12957","DOIUrl":null,"url":null,"abstract":"We provide the optimal receive combining strategy for federated learning in multiple-input multiple-output (MIMO) systems. Our proposed algorithm allows the clients to perform individual gradient sparsification which greatly improves performance in scenarios with heterogeneous (non i.i.d.) training data. The proposed method beats the benchmark by a wide margin.","PeriodicalId":423807,"journal":{"name":"2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"8","resultStr":"{\"title\":\"Optimal MIMO Combining for Blind Federated Edge Learning with Gradient Sparsification\",\"authors\":\"Ema Becirovic, Zheng Chen, E. Larsson\",\"doi\":\"10.48550/arXiv.2203.12957\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We provide the optimal receive combining strategy for federated learning in multiple-input multiple-output (MIMO) systems. Our proposed algorithm allows the clients to perform individual gradient sparsification which greatly improves performance in scenarios with heterogeneous (non i.i.d.) training data. The proposed method beats the benchmark by a wide margin.\",\"PeriodicalId\":423807,\"journal\":{\"name\":\"2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC)\",\"volume\":\"30 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-03-24\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"8\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.48550/arXiv.2203.12957\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.48550/arXiv.2203.12957","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 8

摘要

提出了多输入多输出(MIMO)系统中联邦学习的最优接收组合策略。我们提出的算法允许客户端执行单个梯度稀疏化，这大大提高了在异构(非id)训练数据场景下的性能。所提出的方法大大优于基准方法。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Optimal MIMO Combining for Blind Federated Edge Learning with Gradient Sparsification

We provide the optimal receive combining strategy for federated learning in multiple-input multiple-output (MIMO) systems. Our proposed algorithm allows the clients to perform individual gradient sparsification which greatly improves performance in scenarios with heterogeneous (non i.i.d.) training data. The proposed method beats the benchmark by a wide margin.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

2022 IEEE 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC)

自引率

0.00%

发文量