We Rebuilt a 108-Card Game AI Three Times—From Handcrafted Search to PPO

说明与用途

为GuanDan游戏构建AI需要解决组合动作、信息不完整和延迟信用问题。DanKS实现了两阶段管道：候选人检索和排名，随着版本演进从手工搜索到学习选择，再到使用PPO训练的记忆感知演员-评论家。

问题

Designing an AI for GuanDan, a four-player partnership card game with 108 cards, required handling combinatorial actions, imperfect information, and delayed credit while keeping the action space manageable.

结果

The project demonstrates a clear evolution from handcrafted search to learned candidate selection and finally to a memory‑aware actor‑critic trained with PPO, preserving each generation for research reuse.

工作流

• 1. Decompose the hand and generate a bounded set of legal move candidates.
• 2. Rank the top‑K candidates using a policy that considers visible state, public history, seat context, and structural features.
• 3. In V3, augment the policy with card memory, coverage features, team‑belief signals, and train with PPO and generalized advantage estimation to handle delayed credit.

证据

We built it three times—moving from handcrafted structural search, to learned candidate selection, and finally to a memory-aware actor-critic trained with PPO.

来源

The retrieval stage is not just a speed optimization. Two legal moves may look similar on the table but consume very different parts of the remaining hand.

来源