A Stochastic Proximal-Gradient Method for Regularized Generally Constrained Continuous Optimization
Abstract
A stochastic proximal-gradient method is proposed and analyzed for solving regularized optimization problems subject to nonlinear equality constraints and nonnegativity bounds. The objective function's values and gradients are available only through inexact stochastic oracles. The method decomposes each trial step into one component that reduces linearized infeasibility and a second component that seeks optimality while preserving that reduction. Trial steps are accepted or rejected through a relaxed sufficient decrease condition subject to stochastic merit function evaluations. Without assuming unbiasedness of stochastic gradient estimates, it is proved under reasonable assumptions that the method reaches an ε-KKT point within O(ε−2) iterations with high probability. This guarantee further implies almost-sure termination in finitely many iterations and an O(ε−2) bound on the expected iteration complexity.
Related
- A constraint dissolving method for generally constrained optimization over a nonconvex set
- Follow the Regularized Leader Does Not Converge in Constrained Optimization
- Fast Rates and Strong Convergence of Tikhonov-Regularized Mixed-Order Primal-Dual Dynamics for Linearly Constrained Optimization Without Eventual Ball Conditions
- Regularized extragradient method for structured bilevel optimization in continuous and discrete time
- Reinforcement learning for constrained stochastic linear--quadratic control in continuous time
BibTeX
@misc{curtis2026stochastic,
title = {{A Stochastic Proximal-Gradient Method for Regularized Generally Constrained Continuous Optimization}},
author = {Frank E. Curtis and Xiaoyi Qu and Daniel P. Robinson},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.35752},
eprint = {2610.35752},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.35752},
}
RIS
TY - UNPB
TI - A Stochastic Proximal-Gradient Method for Regularized Generally Constrained Continuous Optimization
AU - Frank E. Curtis
AU - Xiaoyi Qu
AU - Daniel P. Robinson
PY - 2026
DA - 2026/10/26
PB - arXiv
DO - 10.48550/arxiv.2610.35752
UR - https://arxiv.org/abs/2610.35752
AB - A stochastic proximal-gradient method is proposed and analyzed for solving regularized optimization problems subject to nonlinear equality constraints and nonnegativity bounds. The objective function's values and gradients are available only through inexact stochastic oracles. The method decomposes each trial step into one component that reduces linearized infeasibility and a second component that seeks optimality while preserving that reduction. Trial steps are accepted or rejected through a relaxed sufficient decrease condition subject to stochastic merit function evaluations. Without assuming unbiasedness of stochastic gradient estimates, it is proved under reasonable assumptions that the method reaches an ε-KKT point within O(ε^−2) iterations with high probability. This guarantee further implies almost-sure termination in finitely many iterations and an O(ε^−2) bound on the expected iteration complexity.
ER -