Improved Algorithms for Misspecified Linear Markov Decision Processes.
Daniel VialAdvait ParulekarSanjay ShakkottaiR. SrikantPublished in: CoRR (2021)
Keyphrases
- markov decision processes
- policy iteration
- factored mdps
- optimal policy
- learning algorithm
- reinforcement learning
- transition matrices
- reachability analysis
- planning under uncertainty
- state space
- computational complexity
- average cost
- finite state
- action space
- decision processes
- infinite horizon
- reinforcement learning algorithms
- partially observable
- finite horizon
- dynamic programming
- search algorithm