Decentralized Markov Decision Processes with Event-Driven Interactions.
Raphen BeckerShlomo ZilbersteinVictor R. LesserPublished in: AAMAS (2004)
Keyphrases
- markov decision processes
- event driven
- decentralized control
- state space
- finite state
- optimal policy
- decision theoretic planning
- transition matrices
- policy iteration
- reinforcement learning
- dynamic programming
- infinite horizon
- average cost
- planning under uncertainty
- distributed systems
- real time
- model based reinforcement learning
- decision processes
- reinforcement learning algorithms
- state and action spaces
- reachability analysis
- state abstraction
- markov decision process
- real time dynamic programming
- average reward
- action space
- finite horizon
- partially observable
- markup language
- peer to peer
- multi agent
- risk sensitive
- dec pomdps
- factored mdps
- discounted reward
- partially observable markov decision processes
- learning algorithm