Fictitious Play for Mean Field Games: Continuous Time Analysis and Applications

NIPS 2020, 2020.

Cited by: 2|Views48
EI
Weibo:
In this paper we have shown that Fictitious Play can serve as a basis for building practical algorithms to solve a wide variety of Mean Field Games including finite horizon and γ-discounted MFGs as well as games perturbed by a common noise

Abstract:

In this paper, we deepen the analysis of continuous time Fictitious Play learning algorithm to the consideration of various finite state Mean Field Game settings (finite horizon, $\gamma$-discounted), allowing in particular for the introduction of an additional common noise. We first present a theoretical convergence analysis of the c...More

Code:

Data:

0
Your rating :
0

 

Tags
Comments