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

An efficient method for computing leave-one-out error in support vector machines with Gaussian kernels.

Martin M S Lee1, S Sathiya Keerthi, Chong Jin Ong

  • 1Department of Mechanical Engineering, National University of Singapore, Republic of Singapore.

IEEE Transactions on Neural Networks
|September 24, 2004
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

Chemotactic navigation in robotic swimmers via reset-free hierarchical reinforcement learning.

Nature communications·2025
Same author

Distributed Newton Methods for Deep Neural Networks.

Neural computation·2018
Same author

Spiral microchannel with rectangular and trapezoidal cross-sections for size based particle separation.

Scientific reports·2013
Same author

Development and numerical characterization of a new standing wave ultrasonic motor operating in the 30-40kHz frequency range.

Ultrasonics·2013
Same author

Automatic EEG artifact removal: a weighted support vector machine approach with error correction.

IEEE transactions on bio-medical engineering·2009
Same author

Support vector ordinal regression.

Neural computation·2007
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

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

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
Same journal

Data-based identification and control of nonlinear systems via piecewise affine approximation.

IEEE transactions on neural networks·2011
Same journal

Data-core-based fuzzy min-max neural network for pattern classification.

IEEE transactions on neural networks·2011
See all related articles

This study presents an efficient method for calculating leave-one-out (LOO) error in support vector machines (SVMs) with Gaussian kernels. This approach significantly speeds up computation, aiding in hyperparameter tuning and model comparison.

Area of Science:

  • Machine Learning
  • Computational Statistics

Background:

  • Support Vector Machines (SVMs) are powerful classification models.
  • Accurate estimation of leave-one-out (LOO) error is crucial for model evaluation and hyperparameter tuning in SVMs.
  • Standard LOO computation can be computationally intensive, especially for large datasets and complex kernels like the Gaussian kernel.

Purpose of the Study:

  • To develop and present an efficient and accurate method for computing the LOO error of SVMs with Gaussian kernels.
  • To demonstrate the suitability of the proposed method for iterative decomposition techniques used in solving SVMs.
  • To validate the method's performance and highlight its advantages over standard approaches.

Main Methods:

  • An efficient algorithm for computing the LOO error for SVMs with Gaussian kernels was developed.

Related Experiment Videos

  • The method's effectiveness was evaluated using six benchmark datasets.
  • Performance was assessed in the context of iterative decomposition methods for SVM training.

    • Main Results:

    • The proposed method achieves accurate LOO error computation for Gaussian kernel SVMs.
    • Significant speedups, often 10-50 times faster than standard methods, were observed.
    • The method's performance was consistently demonstrated across six diverse benchmark datasets.

    Conclusions:

    • The developed method offers a computationally efficient and accurate way to compute LOO error for Gaussian kernel SVMs.
    • This technique is particularly beneficial for iterative decomposition methods in SVM training.
    • The approach shows strong potential for practical applications in hyperparameter tuning and model selection.