Learning Synergies between Pushing and Grasping with Self-supervised Deep Reinforcement Learning.
Andy ZengShuran SongStefan WelkerJohnny LeeAlberto RodriguezThomas A. FunkhouserPublished in: CoRR (2018)
Keyphrases
- reinforcement learning
- learning process
- learning algorithm
- learning problems
- knowledge acquisition
- learning systems
- deep learning
- function approximation
- learning capabilities
- vision system
- online learning
- supervised learning
- optimal policy
- active learning
- background knowledge
- prior knowledge
- action selection
- reinforcement learning algorithms
- multi agent
- learning agents
- evolutionary learning
- multi agent reinforcement learning
- active exploration