Randomized algorithms and PAC bounds for inverse reinforcement learning in continuous spaces.
Angeliki KamoutsiPeter Schmitt-FörsterTobias SutterVolkan CevherJohn LygerosPublished in: CoRR (2024)
Keyphrases
- randomized algorithms
- inverse reinforcement learning
- lower bound
- upper bound
- approximation algorithms
- preference elicitation
- vc dimension
- sample complexity
- randomized algorithm
- practical problems
- worst case
- reward function
- pac learning
- learning algorithm
- sample size
- constant factor
- multi agent
- generalization error
- search space
- search algorithm