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

Reduced support vector machines: a statistical theory.

Yuh-Jye Lee1, Su-Yun Huang

  • 1Department of Computer Science and Information Engineering, National Taiwan University of Science and Technology, Taipei 106, Taiwan, ROC. yuh-jye@mail.ntust.edu.tw

IEEE Transactions on Neural Networks
|February 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

Machine learning-based monosaccharide profiling for tissue-specific classification of Wolfiporia extensa samples.

Carbohydrate polymers·2023
Same author

Robust Aggregation for Federated Learning by Minimum <i>γ</i>-Divergence Estimation.

Entropy (Basel, Switzerland)·2022
Same author

Alteration of power law scaling of spontaneous brain activity in schizophrenia.

Schizophrenia research·2021
Same author

Radiomic Features at CT Can Distinguish Pancreatic Cancer from Noncancerous Pancreas.

Radiology. Imaging cancer·2021
Same author

Sufficient dimension reduction via random-partitions for the large-p-small-n problem.

Biometrics·2018
Same author

Robust mislabel logistic regression without modeling mislabel probabilities.

Biometrics·2017
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

The reduced support vector machine (RSVM) offers a computationally efficient approach for large datasets by using random subsets. This method maintains model robustness and retains essential information, making it effective for machine learning tasks.

Area of Science:

  • Machine Learning
  • Statistical Learning Theory
  • Computational Statistics

Background:

  • Large datasets pose computational challenges in machine learning.
  • Reduced Support Vector Machines (RSVM) aim to simplify models and improve efficiency.
  • Existing methods may struggle with scalability and model complexity.

Purpose of the Study:

  • To investigate the statistical properties of RSVM, focusing on sampling design and robustness.
  • To analyze the spectral characteristics of the reduced kernel in RSVM.
  • To evaluate the effectiveness of using random subsets for approximating full kernel matrices.

Main Methods:

  • Employing random subset sampling for kernel approximation.
  • Assessing model robustness using criteria like variation, bias, and test power.

Related Experiment Videos

  • Conducting spectral analysis by comparing eigenstructures of full and approximated kernel matrices.
  • Focusing on the statistical theory of reduced set methods within the RSVM context.

    • Main Results:

    • Random subset approximation of kernels retains significant information for learning tasks.
    • The random subset mixture model demonstrates robustness across defined criteria.
    • Small discrepancies between full and approximated kernel eigenstructures validate the approximation.
    • The RSVM approach provides a viable alternative for handling large-scale data.

    Conclusions:

    • RSVM, utilizing random subsets, is a robust and computationally efficient method for large datasets.
    • The spectral analysis confirms that approximated kernels preserve crucial learning information.
    • The random subset approach can enhance existing optimization algorithms.
    • This methodology offers a statistically sound framework for scalable machine learning.