Leave no Trace: Learning to Reset for Safe and Autonomous Reinforcement Learning.
Benjamin EysenbachShixiang GuJulian IbarzSergey LevinePublished in: ICLR (Poster) (2018)
Keyphrases
- reinforcement learning
- autonomous learning
- learning algorithm
- learning process
- learning problems
- learning agents
- prior knowledge
- supervised learning
- learning tasks
- state space
- learning community
- learning systems
- neural network
- reinforcement learning methods
- previously learned
- state abstraction
- multi agent reinforcement learning
- state action
- function approximation
- knowledge acquisition
- online learning
- learning environment
- multi agent
- machine learning