Contrastive Learning as Goal-Conditioned Reinforcement Learning.
Benjamin EysenbachTianjun ZhangSergey LevineRuslan SalakhutdinovPublished in: NeurIPS (2022)
Keyphrases
- reinforcement learning
- learning algorithm
- learning process
- active learning
- knowledge acquisition
- learning problems
- autonomous learning
- neural network
- learning capabilities
- function approximation
- model free
- learning systems
- active exploration
- reinforcement learning methods
- robot control
- markov decision processes
- optimal policy
- unsupervised learning
- decision trees
- machine learning