发表状态 | 已发表Published |
题名 | Federated Multi-Phase Curriculum Learning to Synchronously Correlate User Heterogeneity |
作者 | |
发表日期 | 2024-05-01 |
发表期刊 | IEEE Transactions on Artificial Intelligence
![]() |
ISSN/eISSN | 2691-4581 |
卷号 | 5期号:5页码:2026-2039 |
摘要 | Federated learning (FL) is a decentralized learning method used to train machine learning algorithms. In FL, a global model iteratively collects the parameters of local models without accessing their local data. However, a significant challenge in FL is handling the heterogeneity of local data distribution, which often results in a drifted global model that is difficult to converge. To address this issue, current methods employ different strategies, such as knowledge distillation, weighted model aggregation, and multitask learning. These approaches are referred to as asynchronous FL, as they align user models either locally or posthoc, where model drift has already occurred or has been underestimated. In this article, we propose an active and synchronous correlation approach to address the challenge of user heterogeneity in FL. Specifically, our approach aims to approximate FL as standard deep learning by actively and synchronously scheduling user learning pace in each round with a dynamic multiphase curriculum. A global curriculum is formed by an autoregressive autoencoder that integrates all user curricula on the server. This global curriculum is then divided into multiple phases and broadcast to users to measure and align the domain-agnostic learning pace. Empirical studies demonstrate that our approach outperforms existing asynchronous approaches in terms of generalization performance, even in the presence of severe user heterogeneity. |
关键词 | Curriculum learning (CL) federated learning (FL) heterogeneity data synchronization |
DOI | 10.1109/TAI.2023.3307664 |
URL | 查看来源 |
语种 | 英语English |
Scopus入藏号 | 2-s2.0-85168753609 |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | https://repository.uic.edu.cn/handle/39GCC9TT/11788 |
专题 | 理工科技学院 |
通讯作者 | Guo, Jianxiong |
作者单位 | 1.BNU-HKBU United International College, Guangdong Key Lab of AI and Multi-Modal Data Processing, Department of Computer Science, Zhuhai, 519087, China 2.Beijing Normal University, Advanced Institute of Natural Sciences, Zhuhai, 519087, China 3.BNU-HKBU United International College, Guangdong Key Lab of AI and Multi-Modal Data Processing, Zhuhai, 519087, China |
第一作者单位 | 北师香港浸会大学 |
通讯作者单位 | 北师香港浸会大学 |
推荐引用方式 GB/T 7714 | Wang, Mingjie,Guo, Jianxiong,Jia, Weijia. Federated Multi-Phase Curriculum Learning to Synchronously Correlate User Heterogeneity[J]. IEEE Transactions on Artificial Intelligence, 2024, 5(5): 2026-2039. |
APA | Wang, Mingjie, Guo, Jianxiong, & Jia, Weijia. (2024). Federated Multi-Phase Curriculum Learning to Synchronously Correlate User Heterogeneity. IEEE Transactions on Artificial Intelligence, 5(5), 2026-2039. |
MLA | Wang, Mingjie,et al."Federated Multi-Phase Curriculum Learning to Synchronously Correlate User Heterogeneity". IEEE Transactions on Artificial Intelligence 5.5(2024): 2026-2039. |
条目包含的文件 | 条目无相关文件。 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论