I’m thinking about writing a Julia implementation but I’m a little concerned that the reason I can’t find an implementation is because there are better options. Is there any other approach that you would recommend over CFR?
A couple of updates on this question.
I haven’t been able to find a CFR implementation in Julia, so I’m working on one.
According to Brown, Lerer, Gross, and Sandholm (2018), external-sampling Linear Monte Carlo CFR (Brown and Sandholm 2019) is the state of the art for two player zero sum extensive form games (at least in Poker applications). However, first order methods, such as mirror prox and the excessive gap technique, have faster theoretical convergence and recent results suggest that they can be faster than CFR techniques with appropriate parameter tuning (Kroer, Waugh, Kilinc-Karzan,and Sandholm 2018).