Entropy Maximization for Constrained Markov Decision Processes.
Yagiz SavasMelkior OrnikMurat CubuktepeUfuk TopcuPublished in: Allerton (2018)
Keyphrases
- markov decision processes
- finite state
- state space
- dynamic programming
- optimal policy
- reinforcement learning
- average reward
- transition matrices
- reachability analysis
- decision processes
- policy iteration
- average cost
- reinforcement learning algorithms
- infinite horizon
- risk sensitive
- partially observable
- planning under uncertainty
- decision theoretic planning
- factored mdps
- markov decision process
- finite horizon
- action space
- heuristic search
- action sets
- total reward
- optimal solution
- semi markov decision processes
- objective function