GRIMGEP: Learning Progress for Robust Goal Sampling in Visual Deep Reinforcement Learning.
Grgur KovacAdrien Laversanne-FinotPierre-Yves OudeyerPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- learning process
- learning algorithm
- online learning
- learning problems
- learning systems
- supervised learning
- neural network
- evolutionary learning
- active learning
- prior knowledge
- unsupervised learning
- learning capabilities
- autonomous learning
- deep learning
- robot control
- deep architectures
- model free
- visual information
- knowledge acquisition
- state space
- mobile robot
- dynamic programming
- multi agent