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

Successive overrelaxation for support vector machines.

O L Mangasarian1, D R Musicant

  • 1Computer Sciences Department, University of Wisconsin, Madison, WI 53706, USA.

IEEE Transactions on Neural Networks
|February 7, 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

Nonlinear knowledge in kernel approximation.

IEEE transactions on neural networks·2007
Same author

Importance of nuclear morphology in breast cancer prognosis.

Clinical cancer research : an official journal of the American Association for Cancer Research·1999
Same author

Computer-derived nuclear features compared with axillary lymph node status for breast carcinoma prognosis.

Cancer·1997
Same author

Indeterminate fine-needle aspiration of the breast. Image analysis-assisted diagnosis.

Cancer·1997
Same author

Computer-derived nuclear "grade" and breast cancer prognosis.

Analytical and quantitative cytology and histology·1995
Same author

Computer-derived nuclear features distinguish malignant from benign breast cytology.

Human pathology·1995
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

Successive overrelaxation (SOR) effectively trains support vector machines (SVMs) for massive datasets. This method processes large-scale data efficiently, outperforming other algorithms on big discrimination tasks.

Area of Science:

  • Machine Learning
  • Optimization Algorithms

Background:

  • Training support vector machines (SVMs) on massive datasets presents significant computational challenges.
  • Conventional linear and quadratic programming methods struggle with datasets containing millions of points.

Purpose of the Study:

  • To introduce and evaluate the Successive Overrelaxation (SOR) algorithm for training SVMs on extremely large datasets.
  • To demonstrate SOR's capability in handling data that exceeds available memory.

Main Methods:

  • Utilized Successive Overrelaxation (SOR) for symmetric linear complementarity problems and quadratic programs to train SVMs.
  • Compared SOR's performance against Platt's Sequential Minimal Optimization (SMO) and Joachims' SVMlight algorithms.
  • Applied SOR to datasets with up to 10,000,000 points.

Related Experiment Videos

Main Results:

  • SOR successfully trained SVMs on massive datasets, processing one point at a time, thus not requiring data to reside in memory.
  • The algorithm demonstrated linear convergence to a solution.
  • Numerical results showed SOR's effectiveness on datasets up to 10 million points, solving problems intractable for other methods.
  • On smaller datasets, SOR was faster than SVMlight and comparable or faster than SMO.

Conclusions:

  • Successive Overrelaxation (SOR) offers a viable and efficient solution for training support vector machines on massive datasets.
  • SOR's ability to handle large-scale data makes it a significant advancement for discrimination problems previously unsolvable by conventional techniques.