A Dual Approach to Constrained Markov Decision Processes with Entropy Regularization.
Donghao YingYuhao DingJavad LavaeiPublished in: AISTATS (2022)
Keyphrases
- markov decision processes
- state space
- dynamic programming
- policy iteration
- optimal policy
- transition matrices
- finite state
- reinforcement learning
- decision theoretic planning
- finite horizon
- risk sensitive
- reinforcement learning algorithms
- factored mdps
- reachability analysis
- partially observable
- average cost
- decision processes
- planning under uncertainty
- model based reinforcement learning
- regularization parameter
- machine learning
- action sets
- average reward
- optimal solution
- infinite horizon
- real time dynamic programming
- stochastic shortest path
- multi agent
- decision diagrams