BAIL: Best-Action Imitation Learning for Batch Deep Reinforcement Learning.
Xinyue ChenZijian ZhouZheng WangChe WangYanqiu WuKeith W. RossPublished in: NeurIPS (2020)
Keyphrases
- imitation learning
- reinforcement learning
- action selection
- action space
- reinforcement learning methods
- state space
- function approximation
- mirror neurons
- reinforcement learning algorithms
- learning algorithm
- markov decision processes
- optimal policy
- control problems
- model free
- optimal control
- temporal difference
- learning problems
- supervised learning
- training data
- maximum margin
- control strategies
- multi modal
- markov decision process
- feature space
- feature selection
- computer vision