Design of a Gomoku AI Based on the Alpha-Beta Pruning Search Algorithm

Abstract. This paper presents the implementation of a Gomoku AI based on the alpha-beta pruning search algorithm and the Negamax algorithm. Gomoku, a traditional board game known for its strategic depth, poses significant challenges in AI development due to the exponential increase in possible moves. The AI leverages the computational efficiency of Alpha-Beta Pruning, which enhances the Negamax algorithm by reducing the number of nodes that need to be evaluated in the game tree. This combination allows for faster decision-making without compromising accuracy. Additionally, a value evaluation function is used to assess board states and guide the AI in selecting optimal moves. In the results the performance of the AI was tested through simulations, demonstrating great performance in move selection and computational efficiency compared to traditional methods. The paper also explores potential improvements, including the integration of reinforcement learning (RL) techniques to further enhance the AI's adaptability and strategic decision-making capabilities.