Proximal Reinforcement Learning: Efficient Off-Policy Evaluation in Partially Observed Markov Decision Processes.
Andrew BennettNathan KallusPublished in: Oper. Res. (2024)
Keyphrases
- markov decision processes
- policy evaluation
- reinforcement learning
- policy iteration
- partially observed
- optimal policy
- model free
- least squares
- temporal difference
- state space
- dynamic programming
- finite state
- monte carlo
- reinforcement learning algorithms
- state and action spaces
- markov decision process
- action space
- function approximation
- average reward
- decision processes
- planning under uncertainty
- variance reduction
- stochastic games
- markov decision problems
- reward function
- partially observable markov decision processes
- average cost
- graphical models
- fixed point
- particle filter
- multi agent