English

Making SGD Parameter-Free

Optimization and Control 2024-03-04 v3 Machine Learning Machine Learning

Abstract

We develop an algorithm for parameter-free stochastic convex optimization (SCO) whose rate of convergence is only a double-logarithmic factor larger than the optimal rate for the corresponding known-parameter setting. In contrast, the best previously known rates for parameter-free SCO are based on online parameter-free regret bounds, which contain unavoidable excess logarithmic terms compared to their known-parameter counterparts. Our algorithm is conceptually simple, has high-probability guarantees, and is also partially adaptive to unknown gradient norms, smoothness, and strong convexity. At the heart of our results is a novel parameter-free certificate for SGD step size choice, and a time-uniform concentration result that assumes no a-priori bounds on SGD iterates.

Keywords

Cite

@article{arxiv.2205.02160,
  title  = {Making SGD Parameter-Free},
  author = {Yair Carmon and Oliver Hinder},
  journal= {arXiv preprint arXiv:2205.02160},
  year   = {2024}
}
R2 v1 2026-06-24T11:07:15.731Z