Combining Model-Based Meta-Reasoning and Reinforcement Learning for Adapting Game-Playing Agents.
Patrick UlamJoshua JonesAshok K. GoelPublished in: AIIDE (2008)
Keyphrases
- game playing
- meta reasoning
- temporal difference learning
- reinforcement learning
- opponent modeling
- model free
- multi player
- multi agent
- video games
- case based planning
- general game playing
- multi agent systems
- control knowledge
- game players
- function approximation
- imperfect information
- game tree search
- game play
- action selection
- temporal difference
- open systems
- game tree
- learning outcomes
- dynamic programming
- cooperative
- reinforcement learning algorithms
- computer games
- domain specific
- optimal policy
- attitudes toward
- learning process
- markov decision processes
- autonomous agents
- multiagent systems
- learning experience