M2CURL: Sample-Efficient Multimodal Reinforcement Learning via Self-Supervised Representation Learning for Robotic Manipulation.
Fotios LygerakisVedant DaveElmar RueckertPublished in: UR (2024)
Keyphrases
- reinforcement learning
- learning process
- learning algorithm
- learning problems
- autonomous learning
- supervised learning
- learning systems
- learning capabilities
- perception action
- motor control
- reinforcement learning methods
- robot control
- learning mechanism
- inductive inference
- mobile learning
- sample size
- collaborative learning
- state space
- active learning
- prior knowledge