An Exact Dynamic Programming Solution for a Decentralized Two-Player Markov Decision Process.
Jeff WuSanjay LallPublished in: AAAI Spring Symposium: Embedded Reasoning (2010)
Keyphrases
- markov decision process
- dynamic programming
- state space
- optimal policy
- markov decision processes
- reinforcement learning
- infinite horizon
- exact solution
- multi agent
- dec pomdps
- finite horizon
- policy iteration
- reinforcement learning algorithms
- reward function
- evaluation function
- single agent
- mathematical model
- search algorithm
- cooperative
- optimal solution