Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Automated global structure extraction for effective local building block processing in XCS.

Martin V Butz1, Martin Pelikan, Xavier Llorà

  • 1Department of Cognitive Psychologie, University of Würzburg, Würzburg, 97070, Germany. mbutz@psychologie.uni-wuerzburg.de

Evolutionary Computation
|August 15, 2006
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Looks like SNARC spirit: Coexistence of short- and long-term associations between letters and space.

Quarterly journal of experimental psychology (2006)·2025
Same author

Contextualizing predictive minds.

Neuroscience and biobehavioral reviews·2024
Same author

Infants infer and predict coherent event interactions: Modeling cognitive development.

PloS one·2024
Same author

Virtual reality assessment of a high-calorie food bias: Replication and food-specificity in healthy participants.

Behavioural brain research·2024
Same author

Modal and amodal cognition: an overarching principle in various domains of psychology.

Psychological research·2023
Same author

Active Iterative Social Inference in Multi-Trial Signaling Games.

Open mind : discoveries in cognitive science·2023

This study introduces new crossover operators for Learning Classifier Systems (LCSs), enhancing their ability to solve complex problems by effectively processing feature dependencies. The novel XCS/ECGA and XCS/BOA systems demonstrate improved performance without prior structural knowledge.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Evolutionary Computation

Background:

  • Learning Classifier Systems (LCSs) evolve rule populations for problem-solving.
  • Standard operators in LCSs struggle with problems requiring feature subset processing.
  • Feature interaction disruption leads to poor performance in complex LCS tasks.

Purpose of the Study:

  • Introduce efficient crossover operators for the XCS LCS.
  • Incorporate techniques from Extended Compact Genetic Algorithm (ECGA) and Bayesian Optimization Algorithm (BOA).
  • Develop competent LCSs capable of online dependency structure detection.

Main Methods:

  • Integrated ECGA and BOA crossover mechanisms into the XCS framework.
  • Utilized probabilistic models of the population to generate offspring classifiers.

Related Experiment Videos

  • Evaluated several variations of offspring generation strategies.

    • Main Results:

    • Achieved performance comparable to informed, problem-specific crossover operators.
    • Demonstrated effective online detection and propagation of feature dependency structures.
    • Established the first competent LCSs (XCS/ECGA, XCS/BOA) without prior structural information.

    Conclusions:

    • ECGA and BOA-derived operators significantly improve LCS performance on dependency-heavy problems.
    • Competent LCSs can be developed by integrating advanced GA techniques.
    • Online structure detection and propagation enhance LCS adaptability and efficiency.