Variance Reduction and Low Sample Complexity in Stochastic Optimization via Proximal Point Method

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Liang, Jiaming
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914213175230464
author Liang, Jiaming
author_facet Liang, Jiaming
contents High-probability guarantees in stochastic optimization are often obtained only under strong noise assumptions such as sub-Gaussian tails. We show that such guarantees can also be achieved under the weaker assumption of bounded variance by developing a stochastic proximal point method. This method combines a proximal subproblem solver, which inherently reduces variance, with a probability booster that amplifies per-iteration reliability into high-confidence results. The analysis demonstrates convergence with low sample complexity, without restrictive noise assumptions or reliance on mini-batching.
format Preprint
id arxiv_https___arxiv_org_abs_2402_08992
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Variance Reduction and Low Sample Complexity in Stochastic Optimization via Proximal Point Method
Liang, Jiaming
Optimization and Control
Machine Learning
High-probability guarantees in stochastic optimization are often obtained only under strong noise assumptions such as sub-Gaussian tails. We show that such guarantees can also be achieved under the weaker assumption of bounded variance by developing a stochastic proximal point method. This method combines a proximal subproblem solver, which inherently reduces variance, with a probability booster that amplifies per-iteration reliability into high-confidence results. The analysis demonstrates convergence with low sample complexity, without restrictive noise assumptions or reliance on mini-batching.
title Variance Reduction and Low Sample Complexity in Stochastic Optimization via Proximal Point Method
topic Optimization and Control
Machine Learning
url https://arxiv.org/abs/2402.08992