Model-based reinforcement learning for a multi-player card game with partial observability.
Hajime FujitaShin IshiiPublished in: IAT (2005)
Keyphrases
- game playing
- imperfect information
- partially observable
- markov decision processes
- belief state
- game tree
- partially observable markov decision processes
- perfect information
- video games
- decision problems
- reinforcement learning
- dynamical systems
- state space
- game play
- finite state
- computer games
- learning algorithm
- planning domains
- learning outcomes
- optimal policy
- computational complexity
- markov decision process
- multi agent systems