Online learning for Markov decision processes applied to multi-agent systems.
Mahmoud El ChamieBehçet AçikmeseMehran MesbahiPublished in: CDC (2017)
Keyphrases
- markov decision processes
- online learning
- multi agent systems
- state space
- optimal policy
- reinforcement learning
- dynamic programming
- decision theoretic planning
- finite horizon
- average cost
- transition matrices
- reachability analysis
- finite state
- policy iteration
- planning under uncertainty
- model based reinforcement learning
- reinforcement learning algorithms
- factored mdps
- action space
- decision processes
- cooperative
- multi agent