Dual Parallel Policy Iteration With Coupled Policy Improvement.
Yuhu ChengLongyang HuangC. L. Philip ChenXuesong WangPublished in: IEEE Trans. Neural Networks Learn. Syst. (2024)
Keyphrases
- policy iteration
- markov decision processes
- optimal policy
- model free
- reinforcement learning
- fixed point
- markov decision process
- policy evaluation
- least squares
- average reward
- sample path
- approximate dynamic programming
- infinite horizon
- finite state
- temporal difference
- markov decision problems
- actor critic
- optimal control
- linear programming
- discounted reward
- dynamic programming
- state space
- approximate policy iteration
- long run
- function approximation
- decision problems
- average cost
- convergence rate
- reward function
- sufficient conditions