Stochastic Approximation for Risk-Aware Markov Decision Processes.
Wenjie HuangWilliam B. HaskellPublished in: IEEE Trans. Autom. Control. (2021)
Keyphrases
- stochastic approximation
- markov decision processes
- policy iteration
- risk sensitive
- reinforcement learning
- finite state
- optimal policy
- model free
- state space
- transition matrices
- infinite horizon
- dynamic programming
- average reward
- reinforcement learning algorithms
- markov decision problems
- average cost
- markov decision process
- temporal difference learning
- monte carlo
- partially observable
- least squares
- fixed point
- action space
- stochastic games
- supervised learning