Reversible Markov decision processes and the Gaussian free field.
Venkat AnantharamPublished in: CoRR (2022)
Keyphrases
- markov decision processes
- finite state
- state space
- reinforcement learning
- optimal policy
- planning under uncertainty
- reachability analysis
- dynamic programming
- transition matrices
- decision theoretic planning
- average cost
- markov chain
- risk sensitive
- factored mdps
- policy iteration
- partially observable
- finite horizon
- average reward
- reinforcement learning algorithms
- infinite horizon
- decision processes
- markov decision process
- action space
- reward function
- semi markov decision processes
- model based reinforcement learning
- action sets
- discounted reward