Enable the XCS to dynamically learn multiple problems: A sensor tagging approach

Yao Ming Wu, Po Ming Lee, Liang Yu Chen, Tzu-Chien Hsiao

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

2 Scopus citations

Abstract

The field of presentation of Extended Classifier System(XCS) has undergone many fluctuations and shifts over the years to adapt different domain problems. With the increasing usage of application of artificial intelligence requirements for more complexity presentations of XCS have become more critical. And learning multiple problems is a subject that was needed but few one studied. To dynamically learn multiple problems during a single learning process, we applied a novel representation of classifier conditions of the XCS, named Sensory Tag (ST) to achieve this goal. The XCS with the ST as the input representation is called XCSSTC. The experiments of the proposed method were conducted for the single step problems. The potential of XCSSTC's use in future application of artificial intelligence clearly needs further exploration.

Original languageEnglish
Title of host publicationGECCO 2015 - Companion Publication of the 2015 Genetic and Evolutionary Computation Conference
EditorsSara Silva
PublisherAssociation for Computing Machinery, Inc
Pages1523-1524
Number of pages2
ISBN (Electronic)9781450334884
DOIs
StatePublished - 11 Jul 2015
Event17th Genetic and Evolutionary Computation Conference, GECCO 2015 - Madrid, Spain
Duration: 11 Jul 201515 Jul 2015

Publication series

NameGECCO 2015 - Companion Publication of the 2015 Genetic and Evolutionary Computation Conference

Conference

Conference17th Genetic and Evolutionary Computation Conference, GECCO 2015
Country/TerritorySpain
CityMadrid
Period11/07/1515/07/15

Keywords

  • Learning classifier systems
  • Pattern recognition
  • Scalability
  • XCS

Fingerprint

Dive into the research topics of 'Enable the XCS to dynamically learn multiple problems: A sensor tagging approach'. Together they form a unique fingerprint.

Cite this