JP-DouZero: an enhanced DouDiZhu AI based on reinforcement learning with peasant collaboration and intrinsic rewards.
Mu YuanNikolaos M. FrerisPublished in: BIGCOM (2023)
Keyphrases
- reinforcement learning
- markov decision processes
- artificial intelligence
- machine learning
- expert systems
- function approximation
- state space
- reward function
- information sharing
- case based reasoning
- knowledge representation
- reinforcement learning algorithms
- model free
- learning process
- collaborative learning
- optimal policy
- transfer learning
- reward shaping
- markov decision process
- learning algorithm
- temporal difference
- dynamic programming
- ai technologies
- ai systems
- action selection
- multi agent
- neural network
- knowledge based systems
- intelligent systems
- collaborative environment
- collaborative problem solving
- intelligent behavior
- hidden state
- ai methods
- markov chain
- ai community
- supervised learning
- robotic control
- john mccarthy