English

Maximum Causal Entropy Specification Inference from Demonstrations

Machine Learning 2020-05-19 v5 Robotics Machine Learning

Abstract

In many settings (e.g., robotics) demonstrations provide a natural way to specify tasks; however, most methods for learning from demonstrations either do not provide guarantees that the artifacts learned for the tasks, such as rewards or policies, can be safely composed and/or do not explicitly capture history dependencies. Motivated by this deficit, recent works have proposed learning Boolean task specifications, a class of Boolean non-Markovian rewards which admit well-defined composition and explicitly handle historical dependencies. This work continues this line of research by adapting maximum causal entropy inverse reinforcement learning to estimate the posteriori probability of a specification given a multi-set of demonstrations. The key algorithmic insight is to leverage the extensive literature and tooling on reduced ordered binary decision diagrams to efficiently encode a time unrolled Markov Decision Process. This enables transforming a naive exponential time algorithm into a polynomial time algorithm.

Keywords

Cite

@article{arxiv.1907.11792,
  title  = {Maximum Causal Entropy Specification Inference from Demonstrations},
  author = {Marcell Vazquez-Chanlatte and Sanjit A. Seshia},
  journal= {arXiv preprint arXiv:1907.11792},
  year   = {2020}
}

Comments

Computer Aided Verification, 2020

R2 v1 2026-06-23T10:32:26.178Z