A Dual Approach to Constrained Markov Decision Processes with Entropy Regularization.
Donghao YingYuhao DingJavad LavaeiPublished in: CoRR (2021)
Keyphrases
- markov decision processes
- policy iteration
- dynamic programming
- transition matrices
- optimal policy
- state space
- reinforcement learning
- reachability analysis
- reinforcement learning algorithms
- planning under uncertainty
- infinite horizon
- finite state
- finite horizon
- partially observable
- markov decision process
- decision theoretic planning
- factored mdps
- action space
- average cost
- decision processes
- average reward
- stochastic shortest path
- semi markov decision processes
- state and action spaces
- fixed point