An Object-Oriented Approach to Reinforcement Learning in an Action Game.
Shiwali MohanJohn E. LairdPublished in: AIIDE (2011)
Keyphrases
- reinforcement learning
- agent learns
- extensive form games
- action selection
- reward shaping
- partially observable domains
- state action
- action space
- game theory
- stochastic games
- computer games
- joint action
- state space
- game play
- multi agent
- function approximation
- game design
- learning algorithm
- temporal difference learning
- human actions
- game playing
- video games
- temporal difference
- fitted q iteration
- serious games
- educational games
- markov decision processes
- learning process
- two player games
- human players
- transition model
- reinforcement learning algorithms
- monte carlo tree search
- reasoning about actions
- game development
- initial state
- game tree
- model free
- optimal strategy
- planning problems
- optimal policy