Abstract:Stochastic gradient methods are central to large-scale learning, yet their generalization theory typically relies on independent sampling assumptions. In many practical applications, data are generated by Markov chains and learning is performed in a decentralized manner, which introduces significant analytical challenges. In this work, we investigate the stability and generalization of decentralized stochastic gradient descent (SGD) and stochastic gradient descent ascent (SGDA) under Markov chain sampling. Leveraging a stability-based framework, we characterize how Markovian dependence and decentralized communication jointly influence generalization behavior. Our analysis captures the effects of network topology, Markov chain mixing properties, and primal-dual dynamics. We establish non-asymptotic generalization bounds for both algorithms, extending existing results on Markov stochastic gradient methods to decentralized and minimax settings.
| Comments: | To appear in IJCAI 2026 |
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2605.01701 [cs.LG] |
| (or arXiv:2605.01701v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2605.01701 arXiv-issued DOI via DataCite (pending registration) |
From: Jiahuan Wang [view email]
[v1]
Sun, 3 May 2026 03:58:19 UTC (551 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。