Discount and speed/execution tradeoffs in Markov Decision Process games.
Reinaldo UribeFernando LozanoKatsunari ShibataCharles AndersonPublished in: CIG (2011)
Keyphrases
- markov decision process
- state space
- markov decision processes
- reinforcement learning
- optimal policy
- temporal difference learning
- finite horizon
- transition matrices
- policy iteration
- infinite horizon
- initial state
- video games
- stochastic games
- transition probabilities
- game theory
- machine learning
- markov chain
- stationary policies