Similar articles

Abstract: tibetan Jiu chess, recognized as a national intangible cultural heritage, is a complex game comprising two distinct phases: the layout phase and the battle phase. Improving the performance of deep reinforcement learning (DRL) models for tibetan Jiu chess is challenging, especially given constraints of bhardware resource. To address this, we proposed a two-stage model called JFA, which incorporates hierarchical neural networks and knowledge-guided techniques. The model includes sub-models: strategic layout model (SLM) for the layout phase and hierarchical battle model (HBM) for the battle phase. Both sub-models use similar network structures and employ parallel monte carlo tree search (MCTS) methods for independent self-play training. HBM is structured as a hierarchical neural network, with the upper network selecting movement and jump capture actions, and the lower network handling square capture actions. Human knowledge-based auxiliary agents are introduced to assist SLM and HBM, simulating the entire game and providing reward signals based on square captures or victory outcomes. Additionally, within the HBM model, we propose two human knowledge-based pruning methods that prune parallel MCTS and capture actions in the lower network. In experiments against a layout model utilizing the AlphaZero method, SLM achieved a 74% win rate, with decision times reduced to approximately 147 th of those required by the AlphaZero model. SLM also won first place at 2024 China National Computer Game Tournament. HBM achieved a 70% win rate when playing against other tibetan Jiu chess models. When used together, SLM and HBM in JFA achieved an 81% win rate, comparable to the level of a human amateur 4-dan player. These results demonstrate that JFA efectively enhances arti?cial intelligence (AI) performance in tibetan Jiu chess.