Description
DouZero is an AI system for the card game DouDizhu that enhances traditional Monte-Carlo methods with deep neural networks, action encoding, and parallel actors. The Q-network of DouZero consists of an LSTM to encode historical actions and six layers of MLP with hidden dimension of 512. The network predicts a value for a given state-action pair based on the concatenated representation of action and state.
Papers Using This Method
AlphaDou: High-Performance End-to-End Doudizhu AI Integrating Bidding2024-07-14DouRN: Improving DouZero by Residual Neural Networks2024-03-21DouZero+: Improving DouDizhu AI by Opponent Modeling and Coach-guided Learning2022-04-06DouZero: Mastering DouDizhu with Self-Play Deep Reinforcement Learning2021-06-11