1: \begin{abstract}

2: Counterfactual Regret Minimization (CFR) and its variants are the best algorithms so far for solving large-scale incomplete information games.

3: However,

4: we believe that there are two problems with CFR:

5: First,

6: matrix multiplication is required in CFR iteration,

7: and the time complexity of one iteration is too high;

8: Secondly, the game characteristics in the real world are different.

9: Just using one CFR algorithm will not be perfectly suitable for all game problems.

10: 

11: For these two problems,

12: this paper proposes a new algorithm called Pure CFR (PCFR) based on CFR.

13: PCFR can be seen as a combination of CFR and Fictitious Play (FP),

14: inheriting the concept of counterfactual regret (value) from CFR,

15: and using the best response strategy instead of the regret matching strategy for the next iteration.

16: This algorithm has three advantages.

17: First, PCFR can be combined with any CFR variant.

18: The resulting Pure MCCFR (PMCCFR) can significantly reduce the time and space complexity of one iteration.

19: Secondly,

20: our experiments show that the convergence speed of the PMCCFR is 2$\sim$3 times that of the MCCFR.

21: Finally,

22: there is a type of game that is very suitable for PCFR,

23: we call this type of game clear-game,

24: which is characterized by a high proportion of dominated strategies.

25: Experiments show that in clear-game, the convergence rate of PMCCFR is two orders of magnitude higher than that of MCCFR.

26: \end{abstract}

27: