Complexity guarantees for risk-neutral generalized Nash equilibrium problems
Complexity guarantees for risk-neutral generalized Nash equilibrium problems
In this paper, we address \ac{SGNEP} seeking with risk-neutral agents. Our main contribution lies the development of a stochastic variance-reduced gradient (SVRG) technique, modified to contend with general sample spaces, within a stochastic forward-backward-forward splitting scheme for resolving structured monotone inclusion problems. This stochastic scheme is a double-loop method, in which the mini-batch gradient estimator is computed periodically in the outer loop, while only cheap sampling is required in a frequently activated inner loop, thus achieving significant speed-ups when sampling costs cannot be overlooked. The algorithm is fully distributed and it guarantees almost sure convergence under appropriate batch size and strong monotonicity assumptions. Moreover, it exhibits a linear rate with possible biased estimators, which is rather mild and imposed in many simulation-based optimization schemes. Under monotone regimes, the expectation of the gap function of an averaged iterate diminishes at a suitable sublinear rate while the sample-complexity of computing an $\epsilon$-solution is provably $\mathcal{O}(\epsilon^{-3})$. A numerical study on a class of networked Cournot games reflects the performance of our proposed algorithm.
Haochen Tao、Andrea Iannelli、Meggie Marschner、Mathias Staudigl、Uday V. Shanbhag、Shisheng Cui
数学计算技术、计算机技术
Haochen Tao,Andrea Iannelli,Meggie Marschner,Mathias Staudigl,Uday V. Shanbhag,Shisheng Cui.Complexity guarantees for risk-neutral generalized Nash equilibrium problems[EB/OL].(2025-06-12)[2025-06-23].https://arxiv.org/abs/2506.11409.点此复制
评论