Study on the Play Strategy of Dou Dizhu Poker Based on Convolution Neural Network

Computer game is an important research direction in the field of artificial intelligence. Among them, incomplete information poker games are relatively difficult to implement compared to the intelligent algorithm of chess games. Dou Dizhu game is a kind of incomplete information game poker game. This paper uses convolution neural network, through the analysis large number of game data, to obtain a better landlord identity card model. Experimental results show that the model can provide efficient and reasonable results in Dou Dizhu games, similar to poker played by humans.