Valuation of Convolutional Neural Network in Intelligent Computer Games

Many problems in incomplete information games also exist in our daily life. It is of great significance to solve the problems in our daily life and improve the quality. Texas Holdem is a typical card game with incomplete information. For Texas Holdem a valuation algorithm model based on convolutional neural network is proposed. To train the model using the game history between the masters, to achieve the purpose of learning master skill. The game procedure of the valuation model plays the game with the game procedure designed by the predecessors, and the experimental results show that the convolutional neural network valuation method of learning human master experience can provide better decisions and enhance the Texas Holdem game procedure.