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

Discriminative Sparse Neighbor Approximation for Imbalanced Learning.

Chen Huang, Chen Change Loy, Xiaoou Tang

    IEEE Transactions on Neural Networks and Learning Systems
    |April 1, 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

    XOV-Action: Towards Generalizable Open-Vocabulary Action Recognition.

    IEEE transactions on pattern analysis and machine intelligence·2026
    Same author

    CuDi: Curve Distillation for Efficient and Controllable Exposure Adjustment.

    IEEE transactions on pattern analysis and machine intelligence·2026
    Same author

    LN3Diff++: Scalable Latent Neural Fields Diffusion for Speedy 3D Generation.

    IEEE transactions on pattern analysis and machine intelligence·2025
    Same author

    SMPLest-X: Ultimate Scaling for Expressive Human Pose and Shape Estimation.

    IEEE transactions on pattern analysis and machine intelligence·2025
    Same author

    MeViS: A Multi-Modal Dataset for Referring Motion Expression Video Segmentation.

    IEEE transactions on pattern analysis and machine intelligence·2025
    Same author

    Enhanced Generative Structure Prior for Chinese Text Image Super-Resolution.

    IEEE transactions on pattern analysis and machine intelligence·2025
    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 introduces a new method, discriminative sparse neighbor approximation (DSNA), to address data imbalance in machine learning. DSNA improves prediction accuracy for rare classes, especially in small datasets with overlapping classes.

    Area of Science:

    • Computer Vision
    • Machine Learning
    • Data Science

    Background:

    • Data imbalance is a prevalent issue in vision tasks, leading to biased models favoring majority classes.
    • Conventional methods struggle with imbalanced data, particularly when classes overlap and datasets are small, resulting in poor minority class prediction.
    • Existing approaches often fail to effectively handle severe class imbalance and limited sample sizes.

    Purpose of the Study:

    • To propose a novel method, discriminative sparse neighbor approximation (DSNA), for mitigating class imbalance during prediction.
    • To enhance predictive accuracy for minority classes in imbalanced datasets, including those with significant class overlap.
    • To enable accurate predictions on small datasets by improving extrapolation to unseen data.

    Main Methods:

    Related Experiment Videos

    • Utilizes a cost-sensitive decision forest to identify relevant local training examples for a test sample.
    • Generates class-discriminating, overlapping clusters from the local subset, modeling each as an affine subspace.
    • Employs DSNA to iteratively approximate the test sample within these subspaces, yielding an unbiased prediction.

    Main Results:

    • DSNA effectively mitigates the impact of class imbalance on predictive accuracy.
    • The method demonstrates strong extrapolation capabilities, crucial for accurate predictions on small, imbalanced datasets.
    • Outperforms state-of-the-art methods lacking imbalance handling and rivals deep learning approaches using only hand-crafted features.

    Conclusions:

    • DSNA offers a robust solution for imbalanced learning across classification and regression tasks, regardless of imbalance level.
    • The method's ability to handle class overlap and small sample sizes makes it highly valuable for real-world applications.
    • DSNA presents a significant advancement in imbalanced learning, achieving competitive or superior performance compared to existing techniques.