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

Training SVMs with EDR algorithm.

M B Almeida1, A P Braga, J P Braga

  • 1Department of Electronics Engineering, Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brazil. barros@cpdee.ufmg.br

International Journal of Neural Systems
|September 28, 2001
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

Accurate DMBE potential-energy surface for CNO(<sup>2</sup>A″) and rate coefficients in C(<sup>3</sup>P)+NO collisions.

The Journal of chemical physics·2021
Same author

Radial distribution function for liquid gallium from experimental structure factor: a Hopfield neural network approach.

Journal of molecular modeling·2020
Same author

Fractional kinetics on thermal analysis: application to lumefantrine thermal decomposition.

Journal of molecular modeling·2020
Same author

Quasiclassical Study of the C(<sup>3</sup>P) + NO(X<sup>2</sup>Π) and O(<sup>3</sup>P) + CN(X<sup>2</sup>Σ<sup>+</sup>) Collisional Processes on an Accurate DMBE Potential Energy Surface.

The journal of physical chemistry. A·2019
Same author

A genetic algorithm survey on closed-shell atomic nitrogen clusters employing a quantum chemical approach.

Journal of molecular modeling·2018
Same author

Accurate Explicit-Correlation-MRCI-Based DMBE Potential-Energy Surface for Ground-State CNO.

The journal of physical chemistry. A·2018

This study introduces a novel Support Vector Machine (SVM) training algorithm, Error Dependent Repetition (EDR). EDR prioritizes frequently misclassified patterns, enhancing machine learning efficiency.

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Computer Science

Background:

  • Support Vector Machines (SVMs) are powerful classification tools.
  • Efficient training algorithms are crucial for SVM performance.
  • Existing methods may face challenges with complex datasets.

Purpose of the Study:

  • To introduce a new SVM training algorithm utilizing an Error Dependent Repetition (EDR) strategy.
  • To enhance SVM training efficiency by adapting pattern presentation frequency based on error.
  • To develop an algorithm that solves the SVM dual problem without strict assumptions.

Main Methods:

  • Developed a novel training algorithm for SVMs named SVM-EDR.
  • Implemented an Error Dependent Repetition (EDR) pattern selection strategy.

Related Experiment Videos

  • Utilized a gradient ascent iterative process to solve the dual problem.

    • Main Results:

    • The SVM-EDR algorithm effectively trains SVMs.
    • EDR strategy adapts pattern presentation based on classification error.
    • The algorithm solves the dual problem without assumptions on support vectors or KKT conditions.

    Conclusions:

    • The proposed SVM-EDR algorithm offers an effective approach to SVM training.
    • EDR enhances learning by focusing on difficult patterns.
    • This method provides a flexible alternative for solving SVM dual problems.