Batch reinforcement learning on a RoboCup Small Size League keepaway strategy learning problem

Franco Ollino, Miguel A. Solis, Héctor Allende

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

2 Citations (Scopus)

Abstract

Robotic soccer provides an adversarial scenario where collaborative agents have to execute actions by following a hand-coded or a learned strategy, which in the case of the Small Size League, is given by a centralized decision maker. This work takes advantage of this centralized approach for modelling the keepaway strategy learning problem which is inherently multi-agent, as a single-agent problem, where now each robot forms part of the state of the model. One of the classical reinforcement learning methods is compared with its batch version in terms of amount of time for learning and concluding about updates efficiency based on experiences reusability.

Original languageEnglish
Title of host publicationCRoNe 2018 - Proceedings of the 4th Congress on Robotics and Neuroscience
EditorsCristobal J. Nettle, Miguel A. Solis
PublisherCEUR-WS
ISBN (Electronic)9789560928207
Publication statusPublished - 1 Jan 2018
Externally publishedYes
Event4th Congress on Robotics and Neuroscience, CRoNe 2018 - Valparaiso, Chile
Duration: 15 Nov 201817 Nov 2018

Publication series

NameCEUR Workshop Proceedings
Volume2312
ISSN (Print)1613-0073

Conference

Conference4th Congress on Robotics and Neuroscience, CRoNe 2018
Country/TerritoryChile
CityValparaiso
Period15/11/1817/11/18

ASJC Scopus subject areas

  • Computer Science(all)

Fingerprint

Dive into the research topics of 'Batch reinforcement learning on a RoboCup Small Size League keepaway strategy learning problem'. Together they form a unique fingerprint.

Cite this