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

Input space versus feature space in kernel-based methods.

B Schölkopf1, S Mika, C C Burges

  • 1GMD FIRST, 12489 Berlin, Germany.

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

Outbreak of <i>Serratia marcescens</i> mastitis linked to teat spray contamination in a New Zealand dairy herd.

New Zealand veterinary journal·2026
Same author

Can we estimate herd-level prevalence of lameness in dairy cow herds kept at pasture by sampling part of the herd?

New Zealand veterinary journal·2025
Same author

Repeatability of whole herd lameness scoring: an analysis of a New Zealand dataset.

New Zealand veterinary journal·2024
Same author

Farm-level risk factors and treatment protocols for lameness in New Zealand dairy cattle.

New Zealand veterinary journal·2024
Same author

Prevalence of lameness on pasture-based New Zealand dairy farms: An observational study.

Preventive veterinary medicine·2023
Same author

Investigating the effect of prophylactic claw trimming on the interval between calving and first observed elevated locomotion score in pasture-based dairy cows.

New Zealand veterinary journal·2023
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 explores support vector (SV) kernel feature spaces, detailing their geometry and mapping properties. Algorithms are presented to find input space preimages, improving SV methods and enabling effective nonlinear statistical denoising.

Area of Science:

  • Machine Learning
  • Computational Geometry
  • Kernel Methods

Background:

  • Support Vector (SV) methods rely on kernel functions to map data into high-dimensional feature spaces.
  • Understanding the geometry and properties of these feature spaces is crucial for advancing SV algorithm capacity and performance.
  • Existing research has explored feature space geometry, but connections to input space and practical applications require further investigation.

Purpose of the Study:

  • To advance the understanding of feature spaces associated with support vector (SV) kernel functions.
  • To investigate the geometry of feature spaces and its influence on SV method capacity.
  • To develop methods for finding preimages in input space from feature space vectors and demonstrate their utility.

Main Methods:

Related Experiment Videos

  • Geometric analysis of feature space, including the shape of the input space image under the feature map.
  • Computation of the metric governing the intrinsic geometry of the mapped surface using kernel functions (e.g., inhomogeneous polynomial kernels).
  • Development and application of algorithms to find exact or approximate preimages in input space for given feature space vectors.

Main Results:

  • The study provides insights into the geometric properties of SV kernel feature spaces and their impact on SV method capacity.
  • Algorithms for computing feature space metrics and finding input space preimages are presented.
  • Demonstrated utility of these methods in reducing computational complexity of SV decision functions and in a nonlinear statistical denoising technique using Kernel PCA.

Conclusions:

  • A deeper understanding of SV kernel feature spaces can lead to more efficient and powerful machine learning algorithms.
  • The developed preimage algorithms offer practical benefits for computational efficiency and novel applications like statistical denoising.
  • The findings contribute to the theoretical and applied aspects of kernel methods in pattern recognition and data analysis.