aiqbench analysis
This paper introduces an inexact stochastic variance-reduced gradient (SVRG) method that narrows the complexity gap for computing surely feasible stochastic optimal solutions in constrained convex finite-sum problems. By improving first-order oracle complexity, the approach offers a more efficient alternative to existing penalty methods, particularly when constraint handling is cheaper than objective evaluation.
An inexact stochastic variance-reduced gradient method for constrained convex finite-sum optimization
From arXiv
Full textarXiv:2611.23221v1 Announce Type: cross Abstract: In this paper, we study a class of constrained convex finite-sum optimization problems. In many practical applications, all or some of the constraints are hard constraints that represent imperative requirements. Consequently, any desirable approximate solution must (nearly) satisfy these constraints. Motivated by this, we focus on computing an $\epsilon$-$surely\ feasible\ stochastic\ optimal$ ($\epsilon$-SFSO) solution, in which the constraint violation is deterministically bounded by $\epsilon$, and the expected optimality gap is at most $\epsilon$.
While a recently proposed stochastic variance-reduced first-order penalty method can compute an $\epsilon$-SFSO solution, there remains a significant gap between its first-order oracle complexity for computing an $\epsilon$-SFSO solution and that for computing an $\epsilon$-$expected\ feasible\ stochastic\ optimal$ solution, where both the expected constraint violation and the expected optimality gap are bounded by $\epsilon$. To narrow this gap, we propose an inexact stochastic variance-reduced gradient (SVRG) method that significantly improves the previously best-known first-order oracle complexity for computing an $\epsilon$-SFSO solution.
Furthermore, when the objective consists of a single function, the proposed method reduces to a novel deterministic first-order method for solving standard convex conic optimization. It achieves a significantly lower gradient evaluation complexity with respect to the objective function compared to existing methods, and can outperform them when computing constraint gradients and projections onto the associated cone is cheaper than evaluating the objective gradient. Preliminary numerical results are presented to illustrate the performance of the proposed method.
Abstract:In this paper, we study a class of constrained convex finite-sum optimization problems. In many practical applications, all or some of the constraints are hard constraints that represent imperative requirements. Consequently, any desirable approximate solution must (nearly) satisfy these constraints. Motivated by this, we focus on computing an $\epsilon$-$surely\ feasible\ stochastic\ optimal$ ($\epsilon$-SFSO) solution, in which the constraint violation is deterministically bounded by $\epsilon$, and the expected optimality gap is at most $\epsilon$.
While a recently proposed stochastic variance-reduced first-order penalty method can compute an $\epsilon$-SFSO solution, there remains a significant gap between its first-order oracle complexity for computing an $\epsilon$-SFSO solution and that for computing an $\epsilon$-$expected\ feasible\ stochastic\ optimal$ solution, where both the expected constraint violation and the expected optimality gap are bounded by $\epsilon$. To narrow this gap, we propose an inexact stochastic variance-reduced gradient (SVRG) method that significantly improves the previously best-known first-order oracle complexity for computing an $\epsilon$-SFSO solution.
Furthermore, when the objective consists of a single function, the proposed method reduces to a novel deterministic first-order method for solving standard convex conic optimization. It achieves a significantly lower gradient evaluation complexity with respect to the objective function compared to existing methods, and can outperform them when computing constraint gradients and projections onto the associated cone is cheaper than evaluating the objective gradient. Preliminary numerical results are presented to illustrate the performance of the proposed method.
Optimization and Control (math.OC); Machine Learning (cs.LG)
arXiv:2611.23221 [math.OC]
(or arXiv:2611.23221v1 [math.OC] for this version)
https://doi.org/10.48550/arXiv.2611.23221 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 329 words