Recently, with the rapid development of artificial intelligence technology, there are growing researchers drawing their attention on the field of computer game. In two-player zero-sum extensive games with imperfect information, counterfactual regret (CFR) method is one of the most popular method to solve Nash equilibrium strategy. Therefore, we have carried on a wide range of research and analysis on the CFR and its related improved methods in recent years. In this paper, we firstly introduce the process of solving the strategy of Nash equilibrium based on CFR method. Then, some related improved methods of CFR are reviewed. The bare extended experiments are carried out to help researchers to understand these methods more conveniently. Further, some successful applications and common test platforms are described. Finally, the paper is ended with a conclusion of CFR-based methods and a prediction of future development.

近年来，随着人工智能技术的飞速发展，越来越多的研究人员将注意力转移到计算机游戏领域。在信息不完全的两人零和博弈中，反事实后悔（CFR）方法是解决Nash均衡策略的最流行方法之一。因此，近年来，我们对CFR及其相关的改进方法进行了广泛的研究和分析。本文首先介绍了基于CFR方法求解纳什均衡策略的过程。然后，综述了一些相关的CFR改进方法。进行了裸露的扩展实验，以帮助研究人员更方便地理解这些方法。此外，还介绍了一些成功的应用程序和常见的测试平台。最后，