A Convex Optimization Approach to Distributionally Robust Markov Decision Processes With Wasserstein Distance.
Insoon YangPublished in: IEEE Control. Syst. Lett. (2017)
Keyphrases
- markov decision processes
- convex optimization
- norm minimization
- finite state
- reinforcement learning
- optimal policy
- interior point methods
- transition matrices
- policy iteration
- decision theoretic planning
- dynamic programming
- convex optimization problems
- state space
- average reward
- low rank
- primal dual
- infinite horizon
- total variation
- robust optimization
- average cost
- markov decision process
- multistage
- distance measure
- convex relaxation
- markov decision problems
- higher order