Abstract

We improve robotic learning from demonstration (LfD) via an active learning process of interacting with a human expert to establish a semantic structure and labels for a sign language task. This process situates a learned task in a human-accessible conceptual framework, in order to improve skill transfer not only from expert human teacher to robot, but from robot to novice human learner.

Original languageEnglish (US)
Title of host publicationHRI 2017 - Companion of the 2017 ACM/IEEE International Conference on Human-Robot Interaction
PublisherIEEE Computer Society
Pages271-272
Number of pages2
ISBN (Electronic)9781450348850
DOIs
StatePublished - Mar 6 2017
Event12th Annual ACM/IEEE International Conference on Human-Robot Interaction, HRI 2017 - Vienna, Austria
Duration: Mar 6 2017Mar 9 2017

Publication series

NameACM/IEEE International Conference on Human-Robot Interaction
ISSN (Electronic)2167-2148

Other

Other12th Annual ACM/IEEE International Conference on Human-Robot Interaction, HRI 2017
Country/TerritoryAustria
CityVienna
Period3/6/173/9/17

Keywords

  • hierarchical learning
  • learning from demonstration
  • semantic labels

ASJC Scopus subject areas

  • Artificial Intelligence
  • Human-Computer Interaction
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Teaching and learning using semantic labels'. Together they form a unique fingerprint.

Cite this