Stabilizing Off-Policy Deep Reinforcement Learning from Pixels.
Edoardo CetinPhilip J. BallSteve RobertsOya ÇeliktutanPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- function approximation
- input image
- markov decision processes
- image pixels
- optimal control
- multi agent
- action selection
- nonlinear systems
- temporal difference
- reinforcement learning algorithms
- pixel values
- intensity values
- neighboring pixels
- temporal difference learning
- reinforcement learning methods
- neural network
- optimal policy
- dynamic programming
- data sets
- model free
- pixel wise
- learning capabilities
- homogeneous regions
- pixel classification
- genetic algorithm