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

Learning in the combinatorial neural model.

R J Machado1, V C Barbosa, P A Neves

  • 1Catholic University of Rio de Janeiro, Rio de Janeiro, Brazil.

IEEE Transactions on Neural Networks
|February 8, 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

Pinpointing differentially expressed domains in complex protein mixtures with the cloud service of PatternLab for Proteomics.

Journal of proteomics·2013
Same author

Experimental analysis of lateral impact on planar brittle material: spatial properties of cracks.

Physical review. E, Statistical, nonlinear, and soft matter physics·2011
Same author

Experimental analysis of lateral impact on planar brittle material.

Physical review. E, Statistical, nonlinear, and soft matter physics·2010
Same author

Identifying differences in protein expression levels by spectral counting and feature selection.

Genetics and molecular research : GMR·2008
Same author

Evaluation of human sperm membrane integrity using the water test and the hypoosmotic test.

Andrologia·2001
Same author

Transurethral resection of partially obstructed ejaculatory ducts: seminal parameters and pregnancy outcomes according to the etiology of obstruction.

The Journal of urology·1998
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

This study introduces new algorithms for training combinatorial neural networks (CNM), a type of fuzzy neural network. These methods improve classification and mapping in fuzzy spaces, outperforming existing techniques in real-world applications.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Fuzzy Systems

Background:

  • Combinatorial Neural Models (CNM) are fuzzy neural networks for classification and mapping fuzzy spaces.
  • Learning in CNMs involves complex tasks like optimizing membership functions, network topology, and connection weights.
  • Existing learning algorithms face challenges due to nondifferentiable and nonconvex error functions.

Purpose of the Study:

  • To develop novel algorithms for weight learning in CNMs.
  • To address the complexities arising from nondifferentiable and nonconvex error functions in CNM training.
  • To enhance the performance and robustness of CNMs for classification and mapping tasks.

Main Methods:

  • Introduction of several novel algorithms for CNM weight learning.

Related Experiment Videos

  • Application of subgradient techniques from nondifferentiable optimization.
  • Development of algorithms based on local rules for distributed/parallel implementation.

    • Main Results:

    • Experimental validation on a large-scale satellite image analysis task (Amazon deforestation monitoring).
    • Demonstrated superior performance of a hybrid CNM system compared to error backpropagation techniques.
    • Highlighted the robustness of the hybrid CNM system in practical applications.

    Conclusions:

    • The proposed algorithms offer an effective approach to CNM weight learning.
    • Hybrid CNM systems present a robust and attractive alternative for complex classification and mapping problems.
    • The methods are suitable for distributed and parallel computing environments.