First-order methods for stochastic and finite-sum convex optimization with deterministic constraints

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lu, Zhaosong, Xiao, Yifeng
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915358811619328
author Lu, Zhaosong
Xiao, Yifeng
author_facet Lu, Zhaosong
Xiao, Yifeng
contents In this paper, we study a class of stochastic and finite-sum convex optimization problems with deterministic constraints. Existing methods typically aim to find an $ε$-$expectedly\ feasible\ stochastic\ optimal$ solution, in which the expected constraint violation and expected optimality gap are both within a prescribed tolerance $ε$. However, in many practical applications, constraints must be nearly satisfied with certainty, rendering such solutions potentially unsuitable due to the risk of substantial violations. To address this issue, we propose stochastic first-order methods for finding an $ε$-$surely\ feasible\ stochastic\ optimal$ ($ε$-SFSO) solution, where the constraint violation is deterministically bounded by $ε$ and the expected optimality gap is at most $ε$. Our methods apply an accelerated stochastic gradient (ASG) scheme or a modified variance-reduced ASG scheme $only\ once$ to a sequence of quadratic penalty subproblems with appropriately chosen penalty parameters. We establish first-order oracle complexity bounds for the proposed methods in computing an $ε$-SFSO solution. As a byproduct, we also derive first-order oracle complexity results for sample average approximation method in computing an $ε$-SFSO solution of the stochastic optimization problem using our proposed methods to solve the sample average problem.
format Preprint
id arxiv_https___arxiv_org_abs_2506_20630
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle First-order methods for stochastic and finite-sum convex optimization with deterministic constraints
Lu, Zhaosong
Xiao, Yifeng
Optimization and Control
Machine Learning
Numerical Analysis
In this paper, we study a class of stochastic and finite-sum convex optimization problems with deterministic constraints. Existing methods typically aim to find an $ε$-$expectedly\ feasible\ stochastic\ optimal$ solution, in which the expected constraint violation and expected optimality gap are both within a prescribed tolerance $ε$. However, in many practical applications, constraints must be nearly satisfied with certainty, rendering such solutions potentially unsuitable due to the risk of substantial violations. To address this issue, we propose stochastic first-order methods for finding an $ε$-$surely\ feasible\ stochastic\ optimal$ ($ε$-SFSO) solution, where the constraint violation is deterministically bounded by $ε$ and the expected optimality gap is at most $ε$. Our methods apply an accelerated stochastic gradient (ASG) scheme or a modified variance-reduced ASG scheme $only\ once$ to a sequence of quadratic penalty subproblems with appropriately chosen penalty parameters. We establish first-order oracle complexity bounds for the proposed methods in computing an $ε$-SFSO solution. As a byproduct, we also derive first-order oracle complexity results for sample average approximation method in computing an $ε$-SFSO solution of the stochastic optimization problem using our proposed methods to solve the sample average problem.
title First-order methods for stochastic and finite-sum convex optimization with deterministic constraints
topic Optimization and Control
Machine Learning
Numerical Analysis
url https://arxiv.org/abs/2506.20630