Transfer Learning Heuristically Accelerated Algorithm: A Case Study with Real Robots

Luiz Antonio Celiberto, Reinaldo A.C. Bianchi, Paulo E. Santos

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Reinforcement Learning (RL) is a successful technique for learning the solutions of control problems from an agent's interaction in its domain. However, RL is known to be inefficient for real-world applications. In this paper we propose to use a combination of case-based reasoning (CBR) and heuristically accelerated reinforcement learning methods aiming to speed up a Reinforcement Learning algorithm in a transfer learning problem. We show results of applying this method on a robot soccer domain, where the use of the proposed method led to a significant improvement in the learning rate.

Original languageEnglish
Title of host publicationProceedings - 13th Latin American Robotics Symposium and 4th Brazilian Symposium on Robotics, LARS/SBR 2016
EditorsSergio Vanderlei Cavalcante, Flavio Tonidandel
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages311-316
Number of pages6
ISBN (Electronic)9781509036561
DOIs
Publication statusPublished - 15 Dec 2016
Externally publishedYes
Event13th Latin American Robotics Symposium and 4th Brazilian Symposium on Robotics, LARS/SBR 2016 - Recife, Pernambuco, Brazil
Duration: 8 Oct 201612 Oct 2016

Publication series

NameProceedings - 13th Latin American Robotics Symposium and 4th Brazilian Symposium on Robotics, LARS/SBR 2016

Conference

Conference13th Latin American Robotics Symposium and 4th Brazilian Symposium on Robotics, LARS/SBR 2016
CountryBrazil
CityRecife, Pernambuco
Period8/10/1612/10/16

Keywords

  • Heuristically Accelerated
  • Reinforcement Learning
  • Transfer Learning

Fingerprint Dive into the research topics of 'Transfer Learning Heuristically Accelerated Algorithm: A Case Study with Real Robots'. Together they form a unique fingerprint.

Cite this