Off-Policy Actor-Critic with Emphatic Weightings.
Eric GravesEhsan ImaniRaksha KumaraswamyMartha WhitePublished in: CoRR (2021)
Keyphrases
- actor critic
- reinforcement learning
- optimal control
- policy gradient
- approximate dynamic programming
- temporal difference
- neuro fuzzy
- gradient method
- reinforcement learning algorithms
- policy iteration
- function approximation
- linear program
- step size
- learning algorithm
- recursive least squares
- dynamic programming
- evaluation function
- convergence rate
- partially observable markov decision processes
- markov decision processes
- approximation methods
- average reward
- dynamical systems
- linear programming
- least squares
- objective function