Robust Markov Decision Processes with Data-Driven, Distance-Based Ambiguity Sets.
Sivaramakrishnan RamaniArchis GhatePublished in: SIAM J. Optim. (2022)
Keyphrases
- markov decision processes
- data driven
- finite state
- reinforcement learning
- transition matrices
- state space
- dynamic programming
- optimal policy
- policy iteration
- factored mdps
- risk sensitive
- decision processes
- state and action spaces
- decision theoretic planning
- reachability analysis
- action space
- average reward
- average cost
- partially observable
- reinforcement learning algorithms
- markov decision process
- planning under uncertainty
- probabilistic planning
- state abstraction
- model based reinforcement learning
- semi markov decision processes
- interval estimation
- action sets
- finite horizon
- reward function
- lot sizing
- evaluation function
- least squares