Warm-Start Actor-Critic: From Approximation Error to Sub-optimality Gap.
Hang WangSen LinJunshan ZhangPublished in: ICML (2023)
Keyphrases
- approximation error
- actor critic
- average reward
- reinforcement learning
- policy gradient
- optimal control
- approximate dynamic programming
- temporal difference
- gradient method
- markov decision processes
- neuro fuzzy
- reinforcement learning algorithms
- policy iteration
- optimal policy
- average cost
- long run
- function approximation
- model free
- optimal solution
- action selection
- multiresolution
- finite state
- graph cuts
- supervised learning
- cost function