A Unification of Extensive-Form Games and Markov Decision Processes.
H. Brendan McMahanGeoffrey J. GordonPublished in: AAAI (2007)
Keyphrases
- markov decision processes
- extensive form games
- influence diagrams
- optimal policy
- state space
- reinforcement learning
- transition matrices
- decision problems
- dynamic programming
- finite state
- policy iteration
- planning under uncertainty
- average cost
- partially observable
- higher order
- decision theoretic planning
- infinite horizon
- multi agent
- finite horizon
- reinforcement learning algorithms
- markov decision process
- average reward
- probabilistic inference
- reachability analysis
- model based reinforcement learning
- decision making
- reward function
- factored mdps
- state and action spaces
- special case
- theorem prover
- least squares