Initial State Interventions for Deconfounded Imitation Learning.
Samuel PfrommerYatong BaiHyunin LeeSomayeh SojoudiPublished in: CoRR (2023)
Keyphrases
- initial state
- imitation learning
- reinforcement learning
- situation calculus
- state space
- optimal policy
- robotic systems
- maximum margin
- humanoid robot
- probability distribution
- dynamic programming
- objective function
- markov decision process
- markov chain
- multi modal
- markov decision processes
- reinforcement learning methods
- learning algorithm