English

Continuous-Time Convergence Rates in Potential and Monotone Games

Optimization and Control 2022-02-04 v3 Computer Science and Game Theory Multiagent Systems Systems and Control Systems and Control Dynamical Systems

Abstract

In this paper, we provide exponential rates of convergence to the interior Nash equilibrium for continuous-time dual-space game dynamics such as mirror descent (MD) and actor-critic (AC). We perform our analysis in NN-player continuous concave games that satisfy certain monotonicity assumptions while possibly also admitting potential functions. In the first part of this paper, we provide a novel relative characterization of monotone games and show that MD and its discounted version converge with O(eβt)\mathcal{O}(e^{-\beta t}) in relatively strongly and relatively hypo-monotone games, respectively. In the second part of this paper, we specialize our results to games that admit a relatively strongly concave potential and show AC converges with O(eβt)\mathcal{O}(e^{-\beta t}). These rates extend their known convergence conditions. Simulations are performed which empirically back up our results.

Keywords

Cite

@article{arxiv.2011.10682,
  title  = {Continuous-Time Convergence Rates in Potential and Monotone Games},
  author = {Bolin Gao and Lacra Pavel},
  journal= {arXiv preprint arXiv:2011.10682},
  year   = {2022}
}

Comments

20 pages, 5 figures, manuscript submitted to SIAM Journal on Control and Optimization (SICON) for possible publication

R2 v1 2026-06-23T20:24:31.765Z