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

Between classification-error approximation and weighted least-squares learning.

Kar-Ann Toh1, How-Lung Eng

  • 1Biometrics Engineering Research Center, School of Electrical and Electronic Engineering, Yonsei University, 134 Shinchon-dong, Seodaemun-gu, Seoul, Korea. katoh@yonsei.ac.kr

IEEE Transactions on Pattern Analysis and Machine Intelligence
|February 16, 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

Human Activity Recognition Through Augmented WiFi CSI Signals by Lightweight Attention-GRU.

Sensors (Basel, Switzerland)·2025
Same author

Deep Learning Approach for the Localization and Analysis of Surface Plasmon Scattering.

Sensors (Basel, Switzerland)·2023
Same author

Human Activity Recognition via Score Level Fusion of Wi-Fi CSI Signals.

Sensors (Basel, Switzerland)·2023
Same author

Super-resolved Raman microscopy using random structured light illumination: Concept and feasibility.

The Journal of chemical physics·2021
Same author

Machine learning-based leaky momentum prediction of plasmonic random nanosubstrate.

Optics express·2021
Same author

Surface Plasmon Localization-Based Super-resolved Raman Microscopy.

Nano letters·2020
Same journal

Relation DETR+: Exploring Explicit Position Relation Prior for Dense Prediction.

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

RBF++: Quantifying and Optimizing Reasoning Boundaries across Measurable and Unmeasurable Capabilities for Chain-of-Thought Reasoning.

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

CAFE: Cross-View Adaptive Fusion and Cluster Center Enhancement for Robust Multi-View Clustering.

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

DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving.

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

Ethics-Aware Safe Reinforcement Learning for Rare-Event Risk Control in Interactive Urban Driving.

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

Learning Shape Anchors for Holistic Indoor Scene Understanding.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

This study introduces a novel quadratic approximation for classification error, offering a robust, computationally simple solution superior to traditional least-squares methods for imbalanced data.

Area of Science:

  • Machine Learning
  • Statistical Learning Theory

Background:

  • Traditional least-squares methods can be suboptimal for classification tasks, especially with imbalanced datasets.
  • Existing classification-error based methods may lack computational efficiency or robustness.

Purpose of the Study:

  • To develop a deterministic solution for an approximated classification-error based objective function.
  • To enhance classifier performance and robustness, particularly for unbalanced attribute distributions.

Main Methods:

  • A quadratic approximation of the classification error is proposed for smooth error counting.
  • The solution is linked to weighted least-squares, enabling a robust tuning process.
  • A linear parametric classifier model is adopted for empirical evaluation.

Related Experiment Videos

Main Results:

  • The proposed method demonstrates superiority over the original least-squares-error cost function.
  • The tuning process effectively handles unbalanced attribute distributions.
  • Performance is comparable to state-of-the-art classifiers without compromising computational simplicity.

Conclusions:

  • The novel formulation provides an effective and efficient approach to classification-error based learning.
  • The method offers a robust alternative for handling imbalanced data in classification.
  • This approach balances performance with computational tractability.