Robust Accelerated Dynamics for Subnetwork Bilinear Zero-Sum Games with Distributed Restarting
Optimization and Control
2025-08-07 v2
Abstract
In this paper, we investigate distributed Nash equilibrium seeking for a class of two-subnetwork zero-sum games characterized by bilinear coupling. We present a distributed primal-dual accelerated mirror-descent algorithm with convergence guarantees. However, we demonstrate that this time-varying algorithm is not robust, as it fails to converge under arbitrarily small disturbances. To address this limitation, we introduce a distributed accelerated algorithm that incorporates a coordinated restarting mechanism. We model this new algorithm as a hybrid dynamical system and establish its structural robustness.
Keywords
Cite
@article{arxiv.2504.16835,
title = {Robust Accelerated Dynamics for Subnetwork Bilinear Zero-Sum Games with Distributed Restarting},
author = {Weijian Li and Lacra Pavel and Andreas A. Malikopoulos},
journal= {arXiv preprint arXiv:2504.16835},
year = {2025}
}