Task-embedded control networks for few-shot imitation learning

File Description SizeFormat 
james18.pdfAccepted version6.32 MBAdobe PDFView/Open
Title: Task-embedded control networks for few-shot imitation learning
Authors: James, S
Blosch, M
Davison, A
Item Type: Conference Paper
Abstract: Much like humans, robots should have the ability to leverage knowledge from previously learned tasks in order to learn new tasks quickly in new and unfamiliar environments. Despite this, most robot learning approaches have focused on learning a single task, from scratch, with a limited notion of generalisation, and no way of leveraging the knowledge to learn other tasks more efficiently. One possible solution is meta-learning, but many of the related approaches are limited in their ability to scale to a large number of tasks and to learn further tasks without forgetting previously learned ones. With this in mind, we introduce Task-Embedded Control Networks, which employ ideas from metric learning in order to create a task embedding that can be used by a robot to learn new tasks from one or more demonstrations. In the area of visually-guided manipulation, we present simulation results in which we surpass the performance of a state-of-the-art method when using only visual information from each demonstration. Additionally, we demonstrate that our approach can also be used in conjunction with domain randomisation to train our few-shot learning ability in simulation and then deploy in the real world without any additional training. Once deployed, the robot can learn new tasks from a single real-world demonstration.
Issue Date: 29-Oct-2018
Date of Acceptance: 2-Sep-2018
URI: http://hdl.handle.net/10044/1/64364
Publisher: PMLR
Start Page: 783
End Page: 795
Journal / Book Title: Proceedings of Machine Learning Research
Volume: 87
Copyright Statement: © 2018 by the author(s). Available under a CC-BY Attribution Licence (http://creativecommons.org/licenses/by/4.0)
Sponsor/Funder: Dyson Technology Limited
Funder's Grant Number: PO 4500421830
Conference Name: Conference on Robot Learning
Publication Status: Published
Start Date: 2018-10-29
Finish Date: 2018-10-31
Conference Place: Zurich, Switzerland
Appears in Collections:Faculty of Engineering
Computing



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commons