Intrinsic Reward Driven Imitation Learning via Generative Model.
Xingrui YuYueming LyuIvor W. TsangPublished in: ICML (2020)
Keyphrases
- generative model
- imitation learning
- reinforcement learning
- probabilistic model
- robotic systems
- maximum margin
- prior knowledge
- bayesian framework
- em algorithm
- reward function
- humanoid robot
- posterior probability
- hidden variables
- semi supervised
- topic models
- conditional random fields
- decision trees
- average reward
- learning algorithm
- expectation maximization
- state space
- dynamic programming
- learning process
- feature vectors
- text classification
- markov chain monte carlo
- spatio temporal
- temporal difference
- training data
- probability distribution