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

The Bayesian ARTMAP.

Boaz Vigdor1, Boaz Lerner

  • 1Pattern Analysis and Machine Learning Laboratory, Department of Electrical Computer Engineering, Ben-Gurion University, Beer-Sheva 84105, Israel. boaz@ee.bgu.ac.il

IEEE Transactions on Neural Networks
|December 7, 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

Identifying diagnostic markers in the health records of prediagnostic amyotrophic lateral sclerosis patients.

Amyotrophic lateral sclerosis & frontotemporal degeneration·2025
Same author

Machine learning for predicting Crohn's disease from routine blood tests years before diagnosis: results from the epi-IIRN cohort.

Journal of Crohn's & colitis·2025
Same author

Lower prevalence of metabolic syndrome and its components in the prediagnostic phase of Parkinson's disease.

NPJ Parkinson's disease·2025
Same author

Early Detection of Pulmonary Embolism in a General Patient Population Immediately Upon Hospital Admission Using Machine Learning to Identify New, Unidentified Risk Factors: Model Development Study.

Journal of medical Internet research·2024
Same author

Medical concept embedding of real-valued electronic health records with application to inflammatory bowel disease.

Artificial intelligence in medicine·2023
Same author

Levodopa responsiveness in Parkinson's disease: harnessing real-life experience with machine-learning analysis.

Journal of neural transmission (Vienna, Austria : 1996)·2022
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

Bayesian ARTMAP (BA) enhances fuzzy ARTMAP (FA) neural networks for improved classification accuracy and reduced category proliferation. This Bayesian approach offers better performance across various datasets, outperforming traditional FA methods.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Neuroscience

Background:

  • Fuzzy ARTMAP (FA) is a neural network (NN) known for its classification capabilities.
  • FA can suffer from category proliferation, where too many categories are created, impacting efficiency.
  • Improving classification accuracy while managing category growth is a key challenge in NN research.

Purpose of the Study:

  • To introduce Bayesian ARTMAP (BA), a novel modification of FA using a Bayesian framework.
  • To enhance classification accuracy and reduce category proliferation compared to standard FA.
  • To leverage Bayesian principles for more robust and adaptive category representation and learning.

Main Methods:

  • Developed Bayesian ARTMAP (BA) by integrating a Bayesian framework into FA.

Related Experiment Videos

  • Represented categories using multidimensional Gaussian distributions for flexibility.
  • Implemented adaptive category growth/shrinkage and hypervolume limitation.
  • Utilized Bayes' decision theory for learning, inference, and classification based on minimum expected loss.
  • Estimated class posterior probabilities for probabilistic class prediction.

    • Main Results:

    • BA demonstrated superior classification accuracy over FA on both synthetic and 20 real-world datasets.
    • BA effectively reduced category proliferation compared to FA.
    • The BA model showed improved sensitivity to statistical overlapping and better learning curves.
    • BA achieved lower expected loss, indicating more efficient and accurate classification.

    Conclusions:

    • Bayesian ARTMAP (BA) offers significant improvements over Fuzzy ARTMAP (FA) in classification tasks.
    • The Bayesian framework provides a robust method for enhancing NN performance and managing category complexity.
    • BA's adaptive and probabilistic approach makes it a promising alternative for various classification applications.