Efficient Off-Policy Algorithms for Structured Markov Decision Processes.
Sourav GangulyRaghuram Bharadwaj DiddigiPrabuchandran K. J.Published in: CDC (2023)
Keyphrases
- markov decision processes
- policy iteration
- factored mdps
- dynamic programming
- state space
- transition matrices
- reachability analysis
- reinforcement learning
- finite state
- decision theoretic planning
- planning under uncertainty
- learning algorithm
- reward function
- optimal control
- decision processes
- average reward
- computational complexity
- search algorithm