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

Bayesian support vector regression using a unified loss function.

Wei Chu1, S Sathiya Keerthi, Chong Jin Ong

  • 1Gatsby Computational Neuroscience Unit, University College London, London WC1N 3AR, U.K.

IEEE Transactions on Neural Networks
|September 25, 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

[Research Progress on Mechanism of Phytoremediation for Heavy Metal-contaminated Soil].

Huan jing ke xue= Huanjing kexue·2026
Same author

Genomic and genetic dissection of drought tolerance in a resilient wheat germplasm JIN50.

Nature genetics·2026
Same author

High-Sensitivity Graphene/2D Perovskite Hybrid Photodetector for Visible-Light Sensing on Silicon Nitride Photonic Integrated Platform.

ACS nano·2026
Same author

DC Kerr-driven silicon dual-ring modulator with high linearity, large extinction ratio, and wide bandwidth.

Optics letters·2026
Same author

How meaningfulness and initial learning degree influence associative storage and retrieval processes: A multinomial processing tree and power function analysis.

Psychonomic bulletin & review·2026
Same author

DECR1 deficiency activates a lipid peroxidation-mitocytosis-mitochondrial dysfunction axis in trophoblasts to promote preeclampsia.

Free radical biology & medicine·2026
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 introduces a unified soft insensitive loss function for Bayesian support vector regression, enhancing model adaptation and prediction accuracy. The method offers efficient solutions comparable to traditional support vector regression, even for large datasets.

Area of Science:

  • Machine Learning
  • Statistical Modeling
  • Regression Analysis

Background:

  • Support Vector Regression (SVR) is a powerful regression technique.
  • Bayesian methods offer robust model adaptation and uncertainty quantification.
  • Integrating these approaches can enhance predictive performance.

Purpose of the Study:

  • To introduce a unified soft insensitive loss function for Bayesian Support Vector Regression (BSVR).
  • To demonstrate the benefits of this unified loss function within a Bayesian framework.
  • To evaluate the performance of the proposed BSVR approach.

Main Methods:

  • Developed a unified soft insensitive loss function.
  • Established a Bayesian framework using Gaussian processes for regression.

Related Experiment Videos

  • Utilized the unified loss function in likelihood evaluation.
  • Showcased that maximum a posteriori estimates correspond to an extended SVR problem.

    • Main Results:

    • The proposed BSVR approach combines the merits of SVR (convexity, sparsity) and Bayesian methods (model adaptation, error bars).
    • Experimental results on simulated and real-world data demonstrate effective performance.
    • The method shows efficacy even on large datasets, maintaining computational efficiency.

    Conclusions:

    • The unified soft insensitive loss function provides a robust framework for Bayesian Support Vector Regression.
    • This approach offers a powerful alternative for regression tasks, especially with large datasets.
    • The integration of SVR properties and Bayesian advantages leads to improved regression models.