On the Convergence to a Global Solution of Shuffling-Type Gradient Algorithms
Machine Learning
2023-10-27 v2 Optimization and Control
Abstract
Stochastic gradient descent (SGD) algorithm is the method of choice in many machine learning tasks thanks to its scalability and efficiency in dealing with large-scale problems. In this paper, we focus on the shuffling version of SGD which matches the mainstream practical heuristics. We show the convergence to a global solution of shuffling SGD for a class of non-convex functions under over-parameterized settings. Our analysis employs more relaxed non-convex assumptions than previous literature. Nevertheless, we maintain the desired computational complexity as shuffling SGD has achieved in the general convex setting.
Cite
@article{arxiv.2206.05869,
title = {On the Convergence to a Global Solution of Shuffling-Type Gradient Algorithms},
author = {Lam M. Nguyen and Trang H. Tran},
journal= {arXiv preprint arXiv:2206.05869},
year = {2023}
}
Comments
The 37th Conference on Neural Information Processing Systems (NeurIPS 2023)