Active Incremental Learning of a Contextual Skill Model

Murtaza Hazara, Xiaopu Li, Ville Kyrki

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

74 Downloads (Pure)

Abstract

Contextual skill models are learned to provide skills over a range of task parameters, often using regression across optimal task-specific policies. However, the sequential nature of the learning process is usually neglected. In this paper, we propose to use active incremental learning by selecting a task which maximizes performance improvement over entire task set. The proposed framework exploits knowledge of individual tasks accumulated in a database and shares it among the tasks using a contextual skill model. The framework is agnostic to the type of policy representation, skill model, and policy search. We evaluated the skill improvement rate in two tasks, ball-in-a-cup and basketball. In both, active selection of tasks lead to a consistent improvement in skill performance over a baseline.
Original languageEnglish
Title of host publicationProceedings of the 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2019
PublisherIEEE
Pages1834-1839
Number of pages6
ISBN (Electronic)978-1-7281-4004-9
DOIs
Publication statusPublished - 2019
MoE publication typeA4 Article in a conference publication
EventIEEE/RSJ International Conference on Intelligent Robots and Systems - The Venetian Macao, Macau, China
Duration: 4 Nov 20198 Nov 2019
https://www.iros2019.org/

Publication series

NameProceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems
PublisherIEEE
ISSN (Print)2153-0858
ISSN (Electronic)2153-0866

Conference

ConferenceIEEE/RSJ International Conference on Intelligent Robots and Systems
Abbreviated titleIROS
CountryChina
CityMacau
Period04/11/201908/11/2019
Internet address

Keywords

  • Active incrementl learning
  • Contextual skill model
  • Task parameters
  • Optima task-specific policies
  • Learning process
  • Skill improvement rate
  • Skill performance
  • Policy representation
  • Policy search
  • Ball-in-a-cup
  • Basketball
  • Learning (artificial intelligence)

Fingerprint Dive into the research topics of 'Active Incremental Learning of a Contextual Skill Model'. Together they form a unique fingerprint.

Cite this