Optimization of the randomness in Einstein which based on Monte Carlo Algorithms

As an incomplete information game, Einstein's unique process of rolling the dice brings great randomness. The randomness of its chess moving makes it difficult to apply direct reference to other relatively mature chess algorithms. With the accumulation and development of competition, the existing Einstein program has already had preliminary intelligence. The existing Einstein algorithm is mainly based on Minimax algorithm with static evaluation and Monte Carlo with pruning algorithm. However, its large randomness and limitation of pruning constrain improving. This paper which based on Monte Carlo analyses the evaluation and its corresponding simulation. From the perspective of the data, we can explore the advantages and disadvantages of Monte Carlo and improve the algorithm to judge the current situation and choose the best move, so that providing different ideas of pruning to optimize Monte Carlo program.