English

Asymptotic Time-Uniform Inference for Parameters in Averaged Stochastic Approximation

Machine Learning 2024-10-22 v1 Machine Learning Methodology

Abstract

We study time-uniform statistical inference for parameters in stochastic approximation (SA), which encompasses a bunch of applications in optimization and machine learning. To that end, we analyze the almost-sure convergence rates of the averaged iterates to a scaled sum of Gaussians in both linear and nonlinear SA problems. We then construct three types of asymptotic confidence sequences that are valid uniformly across all times with coverage guarantees, in an asymptotic sense that the starting time is sufficiently large. These coverage guarantees remain valid if the unknown covariance matrix is replaced by its plug-in estimator, and we conduct experiments to validate our methodology.

Keywords

Cite

@article{arxiv.2410.15057,
  title  = {Asymptotic Time-Uniform Inference for Parameters in Averaged Stochastic Approximation},
  author = {Chuhan Xie and Kaicheng Jin and Jiadong Liang and Zhihua Zhang},
  journal= {arXiv preprint arXiv:2410.15057},
  year   = {2024}
}

Comments

35 pages, 4 figures

R2 v1 2026-06-28T19:28:11.975Z