Noisy Importance Sampling Actor-Critic: An Off-Policy Actor-Critic With Experience Replay.
Norman L. TasfiMiriam A. M. CapretzPublished in: IJCNN (2020)
Keyphrases
- actor critic
- importance sampling
- monte carlo
- temporal difference
- reinforcement learning
- optimal control
- policy gradient
- approximate dynamic programming
- neuro fuzzy
- gradient method
- reinforcement learning algorithms
- variance reduction
- markov chain
- function approximation
- particle filter
- kalman filter
- policy iteration
- average reward
- step size
- evaluation function
- state space
- learning algorithm
- model free
- approximate inference
- genetic algorithm
- particle filtering
- dynamic programming