A Unified Linear Programming Framework for Offline Reward Learning from Human Demonstrations and Feedback.
Kihyun KimJiawei ZhangAsuman E. OzdaglarPablo A. ParriloPublished in: CoRR (2024)
Keyphrases
- linear programming
- reinforcement learning
- learning algorithm
- learning scheme
- prior knowledge
- unified model
- learning systems
- learning problems
- learning process
- real time
- motor skills
- unsupervised learning
- language acquisition
- theoretical framework
- human teacher
- dynamic bayesian networks
- linear program
- main contribution
- online learning
- supervised learning
- np hard