Stochastic Weakly Convex Optimization Beyond Lipschitz Continuity
Optimization and Control
2024-11-07 v2 Machine Learning
Abstract
This paper considers stochastic weakly convex optimization without the standard Lipschitz continuity assumption. Based on new adaptive regularization (stepsize) strategies, we show that a wide class of stochastic algorithms, including the stochastic subgradient method, preserve the convergence rate with constant failure rate. Our analyses rest on rather weak assumptions: the Lipschitz parameter can be either bounded by a general growth function of or locally estimated through independent random samples.
Cite
@article{arxiv.2401.13971,
title = {Stochastic Weakly Convex Optimization Beyond Lipschitz Continuity},
author = {Wenzhi Gao and Qi Deng},
journal= {arXiv preprint arXiv:2401.13971},
year = {2024}
}