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

Scaling genetic programming to large datasets using hierarchical dynamic subset selection.

Robert Curry, Peter Lichodzijewski, Malcolm I Heywood

    IEEE Transactions on Systems, Man, and Cybernetics. Part B, Cybernetics : a Publication of the IEEE Systems, Man, and Cybernetics Society
    |August 19, 2007
    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

    Warmer autumns and winters could reduce honey bee overwintering survival with potential risks for pollination services.

    Scientific reports·2024
    Same author

    Inferring pesticide toxicity to honey bees from a field-based feeding study using a colony model and Bayesian inference.

    Ecological applications : a publication of the Ecological Society of America·2021
    Same author

    Emergent Solutions to High-Dimensional Multitask Reinforcement Learning.

    Evolutionary computation·2018
    Same author

    Counterfeit reagents: Fake serum has telltale fingerprint.

    Nature·2017
    Same author

    Population growth of Varroa destructor (Acari: Varroidae) in commercial honey bee colonies treated with beta plant acids.

    Experimental & applied acarology·2014
    Same author

    Classification as clustering: a Pareto cooperative-competitive GP approach.

    Evolutionary computation·2010
    Same journal

    Strategic Ability Updating in Concurrent Games by Coalitional Commitment.

    IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2015
    Same journal

    Meta-Analysis of the First Facial Expression Recognition Challenge.

    IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2012
    Same journal

    Adjustable model-based fusion method for multispectral and panchromatic images.

    IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2012
    Same journal

    Face Feature Weighted Fusion Based on Fuzzy Membership Degree for Video Face Recognition.

    IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2012
    Same journal

    A New Adaptive Fast Cellular Automaton Neighborhood Detection and Rule Identification Algorithm.

    IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2012
    Same journal

    Human-arm-and-hand-dynamic model with variability analyses for a stylus-based haptic interface.

    IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2012
    See all related articles

    New active learning algorithms, including cascaded and balanced block dynamic subset selection (DSS), significantly improve genetic programming (GP) classification accuracy and reduce training time without hardware. These methods enhance GP performance efficiently.

    Area of Science:

    • Computational intelligence and machine learning.
    • Optimization algorithms and evolutionary computation.

    Background:

    • Genetic programming (GP) faces computational overhead challenges, impacting its practical application.
    • Active learning, specifically subset selection heuristics like random (RSS) and dynamic (DSS), offers a software-based solution.

    Discussion:

    • Introduces novel hierarchical dynamic subset selection (DSS) algorithms: RSS-DSS, cascaded RSS-DSS, and the previously undescribed balanced block DSS.
    • Evaluates these algorithms on large-scale, unbalanced real-world binary classification problems (30,000–500,000 exemplars).
    • Compares performance against GP without active learning and the original RSS-DSS algorithm.

    Key Insights:

    • The cascaded and balanced block DSS algorithms significantly enhance classification accuracy compared to RSS-DSS.

    Related Experiment Videos

  • These advanced DSS methods reduce the occurrence of degenerate solutions in GP.
  • GP training time is drastically reduced from hours to minutes without compromising classification performance.

    • Outlook:

    • These active learning strategies provide a viable approach to mitigate GP's computational demands.
    • Potential for broader application in complex machine learning tasks requiring efficient training.
    • Further research into adaptive and hierarchical subset selection for evolutionary algorithms.