Continuous-Time Convergence Rates in Potential and Monotone Games

11/21/2020 ∙ by Bolin Gao, et al. ∙ 0 ∙

In this paper, we provide exponential rates of convergence to the Nash equilibrium of continuous-time game dynamics such as mirror descent (MD) and actor-critic (AC) in N-player continuous games that are either potential games or monotone games but possibly potential-free. In the first part of this paper, under the assumption the game admits a relatively strongly concave potential, we show that MD and AC converge in 𝒪(e^-β t). In the second part of this paper, using relative concavity, we provide a novel relative characterization of monotone games and show that MD and its discounted version converge with 𝒪(e^-β t) in relatively strongly and relatively hypo-monotone games. Moreover, these rates extend their known convergence conditions and also improve the results in the potential game setup. Simulations are performed which empirically back up our results.



There are no comments yet.


page 1

page 2

page 3

page 4

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.