Beyond the Policy Gradient Theorem for Efficient Policy Updates in Actor-Critic Algorithms.
Romain LarocheRemi TachetPublished in: CoRR (2022)
Keyphrases
- policy gradient
- actor critic
- policy gradient methods
- reinforcement learning
- policy iteration
- partially observable markov decision processes
- function approximation
- approximation methods
- optimal control
- approximate dynamic programming
- learning algorithm
- reinforcement learning algorithms
- gradient method
- reinforcement learning methods
- temporal difference
- natural actor critic
- function approximators
- neuro fuzzy
- average reward
- markov decision processes
- state space