CaltechAUTHORS
  A Caltech Library Service

Neural Task Programming: Learning to Generalize Across Hierarchical Tasks

Xu, Danfei and Nair, Suraj and Zhu, Yuke and Gao, Julian and Garg, Animesh and Fei-Fei, Li and Savarese, Silvio (2018) Neural Task Programming: Learning to Generalize Across Hierarchical Tasks. In: 2018 IEEE International Conference on Robotics and Automation (ICRA). IEEE , Piscataway, NJ, pp. 3795-3802. ISBN 978-1-5386-3081-5. http://resolver.caltech.edu/CaltechAUTHORS:20180921-091302880

Full text is not posted in this repository. Consult Related URLs below.

Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:20180921-091302880

Abstract

In this work, we propose a novel robot learning framework called Neural Task Programming (NTP), which bridges the idea of few-shot learning from demonstration and neural program induction. NTP takes as input a task specification (e.g., video demonstration of a task) and recursively decomposes it into finer sub-task specifications. These specifications are fed to a hierarchical neural program, where bottom-level programs are callable subroutines that interact with the environment. We validate our method in three robot manipulation tasks. NTP achieves strong generalization across sequential tasks that exhibit hierarchal and compositional structures. The experimental results show that NTP learns to generalize well towards unseen tasks with increasing lengths, variable topologies, and changing objectives.stanfordvl.github.io/ntp/.


Item Type:Book Section
Related URLs:
URLURL TypeDescription
https://doi.org/10.1109/ICRA.2018.8460689DOIArticle
https://ieeexplore.ieee.org/document/8460689PublisherArticle
Additional Information:© 2018 IEEE. This research was performed at the SVL at Stanford in affiliation with the Stanford AI Lab, Stanford-Toyota AI Center.
Record Number:CaltechAUTHORS:20180921-091302880
Persistent URL:http://resolver.caltech.edu/CaltechAUTHORS:20180921-091302880
Official Citation:D. Xu et al., "Neural Task Programming: Learning to Generalize Across Hierarchical Tasks," 2018 IEEE International Conference on Robotics and Automation (ICRA), Brisbane, Australia, 2018, pp. 1-8. doi: 10.1109/ICRA.2018.8460689
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:89821
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:21 Sep 2018 16:24
Last Modified:21 Sep 2018 16:24

Repository Staff Only: item control page