Acceleration of Actor-Critic Deep Reinforcement Learning for Visual Grasping in Clutter by State Representation Learning Based on Disentanglement of a Raw Input Image.
Tae Won KimYeseong ParkYoungbin ParkIl Hong SuhPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- actor critic
- input image
- state space
- learning algorithm
- reinforcement learning algorithms
- policy gradient
- learning process
- temporal difference
- optimal control
- policy gradient methods
- dynamic programming
- gradient method
- learning problems
- neuro fuzzy
- function approximation
- learning tasks
- action selection
- partially observable
- state action
- reinforcement learning methods
- multi agent
- dynamical systems
- action space
- multiscale
- machine learning