Unsupervised Perceptual Rewards for Imitation Learning.
Pierre SermanetKelvin XuSergey LevinePublished in: ICLR (Workshop) (2017)
Keyphrases
- imitation learning
- reinforcement learning
- robotic systems
- unsupervised learning
- maximum margin
- supervised learning
- humanoid robot
- markov decision processes
- semi supervised
- machine learning
- state space
- function approximation
- computer vision
- learning algorithm
- pairwise
- reinforcement learning algorithms
- control problems
- reinforcement learning methods