---
title: "Automatic State Abstraction from Demonstration"
authors:
  - "Luis C. Cobo"
  - "Peng Zang"
  - "Charles Lee Isbell Jr."
  - "Andrea Lockerd Thomaz"
doi: "10.5591/978-1-57735-516-8/IJCAI11-211"
url: "https://ijcai.org/proceedings/2011/211"
pdf: "https://ijcai.org/proceedings/2011/0211.pdf"
venue: "Proceedings of the Twenty-Second International Joint Conference on Artificial Intelligence"
year: 2011
pages: "1243-1247"
track: ""
keywords:
---

# Automatic State Abstraction from Demonstration

Luis C. Cobo, Peng Zang, Charles Lee Isbell Jr., Andrea Lockerd Thomaz

*Proceedings of the Twenty-Second International Joint Conference on Artificial Intelligence*, pages 1243-1247. DOI: [10.5591/978-1-57735-516-8/IJCAI11-211](https://doi.org/10.5591/978-1-57735-516-8/IJCAI11-211)

## Abstract

Learning from Demonstration (LfD) is a popular technique for building decision-making agents from human help. Traditional LfD methods use demonstrations as training examples for supervised learning, but complex tasks can require more examples than is practical to obtain. We present Abstraction from Demonstration (AfD), a novel form of LfD that uses demonstrations to infer state abstractions and reinforcement learning (RL) methods in those abstract state spaces to build a policy. Empirical results show that AfD is greater than an order of magnitude more sample efficient than jus tusing demonstrations as training examples, and exponentially faster than RL alone.