Gap-Dependent Unsupervised Exploration for Reinforcement Learning.
Jingfeng WuVladimir BravermanLin YangPublished in: AISTATS (2022)
Keyphrases
- reinforcement learning
- active exploration
- exploration strategy
- action selection
- supervised learning
- exploration exploitation
- unsupervised learning
- model based reinforcement learning
- function approximation
- semi supervised
- reinforcement learning algorithms
- state space
- data driven
- autonomous learning
- learning algorithm
- learning process
- exploration exploitation tradeoff
- temporal difference
- supervised classification
- markov decision processes
- transfer learning
- neural network
- model free
- unsupervised manner
- function approximators
- temporal difference learning
- reinforcement learning methods
- dynamic programming
- active learning
- robotic control
- multi agent