Average-Reward Decentralized Markov Decision Processes.
Marek PetrikShlomo ZilbersteinPublished in: IJCAI (2007)
Keyphrases
- average reward
- markov decision processes
- optimal policy
- policy iteration
- discounted reward
- semi markov decision processes
- optimality criterion
- state space
- dynamic programming
- finite state
- stochastic games
- reinforcement learning
- multi agent
- partially observable
- reinforcement learning algorithms
- long run
- dec pomdps
- decision theoretic planning
- factored mdps
- state and action spaces
- infinite horizon
- action space
- decision processes
- average cost
- total reward
- planning under uncertainty
- hierarchical reinforcement learning
- markov decision process
- action sets