Towards Safe Policy Improvement for Non-Stationary MDPs.
Yash ChandakScott M. JordanGeorgios TheocharousMartha WhitePhilip S. ThomasPublished in: CoRR (2020)
Keyphrases
- non stationary
- finite horizon
- optimal policy
- markov decision processes
- markov decision process
- markov decision problems
- policy search
- policy iteration
- infinite horizon
- adaptive algorithms
- average reward
- partially observable
- reinforcement learning
- state and action spaces
- dynamic programming
- factored mdps
- average cost
- reward function
- autoregressive
- long run
- white noise
- change point detection
- reinforcement learning problems
- finite state
- decision processes
- temporal evolution
- action space
- partially observable markov decision processes
- empirical mode decomposition
- stock price
- linear programming
- state space
- biomedical signals