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

On the Impact of Regularization Variation on Localized Multiple Kernel Learning.

Yina Han, Kunde Yang, Yixin Yang

    IEEE Transactions on Neural Networks and Learning Systems
    |April 20, 2017
    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

    Freestanding ferroelectric membranes via ionic unlocking van der Waals interface.

    Nature communications·2026
    Same author

    A mechano-integrated gradient electrolyte for long-cycling solid-state lithium metal batteries.

    Nature communications·2026
    Same author

    Bearings-only acoustic source localization method using two distributed gliders and deep ocean experimental validation in the South China Sea.

    JASA express letters·2026
    Same author

    A novel carbazole-type ratiometric fluorescent probe from natural nopinone for ultrasensitive and visual detection of BPO and its application in food and cosmetic samples.

    Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy·2026
    Same author

    Advanced therapy treatment patterns in moderate-to-severe ulcerative colitis: a long-term retrospective claims analysis.

    Crohn's & colitis 360·2026
    Same author

    A multi-array bearing-only fusion framework for passive underwater multi-target localization.

    The Journal of the Acoustical Society of America·2026
    Same journal

    Hidden Data Recovery and Forecasting via Next-Generation Reservoir Computing With Multiscale Delay Selection.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    CAFF-CIL: Causality-Aware Freedom Forgetting Approach for Class-Incremental Learning.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    Harmonic Autoencoding Framework for Multiple Tasks in Magnetic Particle Imaging Reconstruction.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    A Survey on Human-Centric Voice-Face Multimodal Learning.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    Vision-Assisted Foundation Model for Solving Multitask Vehicle Routing Problems.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    FP3O: Enabling Proximal Policy Optimization in Multiagent Cooperation With Parameter-Sharing Versatility.

    IEEE transactions on neural networks and learning systems·2026
    See all related articles

    This study examines how different regularization methods for localized kernel weights impact localized multiple kernel learning (LMKL) algorithms. Matrix-regularized LMKL demonstrates superior performance compared to vector and samplewise regularizations.

    Area of Science:

    • Machine Learning
    • Computational Statistics

    Background:

    • Localized Multiple Kernel Learning (LMKL) algorithms utilize localized kernel weights, with various regularization techniques proposed.
    • Different regularization strategies lead to diverse LMKL formulations and solution approaches.

    Purpose of the Study:

    • To analyze the impact of regularization variations on localized kernel weights within LMKL algorithms.
    • To provide stability bounds for three LMKL methods based on the norm of localized kernel weight variations.

    Main Methods:

    • Applied stability analysis theory (Bousquet and Elisseeff) to derive stability bounds.
    • Investigated three LMKL methods: vector -norm LMKL, matrix-regularized -norm LMKL, and samplewise -norm LMKL within a support vector machine framework.
    • Conducted comparative analysis of derived stability bounds to infer performance differences.

    Related Experiment Videos

    Main Results:

    • Derived theoretical stability bounds for the three LMKL regularization methods.
    • Qualitatively revealed performance differences based on stability bounds: matrix-regularized LMKL > vector -norm LMKL > samplewise -norm LMKL.
    • Empirically validated theoretical findings through experiments on ten UCI benchmark datasets.

    Conclusions:

    • Regularization of localized kernel weights significantly influences LMKL algorithm performance.
    • Matrix-regularized LMKL offers superior stability and performance.
    • Theoretical analysis aligns with empirical results, confirming the effectiveness of the proposed stability bounds.