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

Constructive feedforward ART clustering networks. II.

A Baraldi1, E Alpaydin

  • 1ICSI, Berkeley, CA.

IEEE Transactions on Neural Networks
|February 5, 2008
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

Hydrocarbon decomposition kinetics on the Ir(111) surface.

Physical chemistry chemical physics : PCCP·2018
Same author

Theoretical and experimental investigation of optical absorption anisotropy in β-Ga2O3.

Journal of physics. Condensed matter : an Institute of Physics journal·2016
Same author

Interplay between electronic and structural properties in the Pb/Ag(1 0 0) interface.

Journal of physics. Condensed matter : an Institute of Physics journal·2015
Same author

Human nail plate modifications induced by onychomycosis: implications for topical therapy.

Pharmaceutical research·2014
Same author

Fine tuning of graphene-metal adhesion by surface alloying.

Scientific reports·2013
Same author

Surface structure of Pd(111) with less than half a monolayer of Zn.

Physical chemistry chemical physics : PCCP·2013
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

A new clustering algorithm, fully self-organizing SART (FOSART), is introduced for unsupervised learning. FOSART offers a balanced approach to clustering tasks, enhancing user interaction and accuracy.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Neuroscience

Background:

  • The study builds upon prior work defining constructive unsupervised on-line learning simplified adaptive resonance theory (SART) clustering networks.
  • Previous instances, Symmetric Fuzzy ART (S-Fuzzy ART) and Gaussian ART (GART), are established.
  • This paper introduces a novel network within the SART class.

Purpose of the Study:

  • To present and discuss a new SART network, termed fully self-organizing SART (FOSART).
  • To detail FOSART's capabilities as a constructive, soft-to-hard competitive, topology-preserving, minimum-distance-to-means clustering algorithm.
  • To evaluate FOSART's performance against existing clustering techniques.

Main Methods:

  • FOSART is designed as a constructive algorithm that generates and removes processing units and lateral connections dynamically.

Related Experiment Videos

  • It employs a minimum-distance-to-means clustering approach.
  • FOSART was compared with Fuzzy ART, S-Fuzzy ART, GART, neural gas, and self-organizing maps.

    • Main Results:

    • FOSART demonstrated effective performance in unsupervised learning tasks including vector quantization, perceptual grouping, and 3-D surface reconstruction.
    • The algorithm proved capable of both generating and removing network components adaptively.
    • Comparative experiments highlighted FOSART's strengths relative to other methods.

    Conclusions:

    • FOSART represents a significant advancement in SART clustering networks.
    • The algorithm achieves a favorable balance of user-friendliness, accuracy, efficiency, robustness, and flexibility.
    • FOSART shows promise for various complex unsupervised learning applications.