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

Robust boosting algorithm against mislabeling in multiclass problems.

Takashi Takenouchi1, Shinto Eguchi, Noboru Murata

  • 1Nara Institute of Science and Technology, Graduate School of Information Science, Ikoma, Nara 630-0192, Japan. ttakashi@is.naist.jp

Neural Computation
|January 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

Noiseless Diffusion-GAN: Scaling-based data augmentation for generative models.

Neural networks : the official journal of the International Neural Network Society·2025
Same author

Tensor dictionary-based heterogeneous transfer learning to study emotion-related gender differences in brain.

Neural networks : the official journal of the International Neural Network Society·2024
Same author

Denoising cosine similarity: A theory-driven approach for efficient representation learning.

Neural networks : the official journal of the International Neural Network Society·2024
Same author

Learning domain invariant representations by joint Wasserstein distance minimization.

Neural networks : the official journal of the International Neural Network Society·2023
Same author

Deep Clustering With a Constraint for Topological Invariance Based on Symmetric InfoNCE.

Neural computation·2023
Same author

[Effectiveness of Prophylactic Negative Pressure Wound Therapy after Cardiovascular Surgery through Median Sternotomy].

Kyobu geka. The Japanese journal of thoracic surgery·2022
Same journal

A Model-Free Reinforcement Learning Implementation of Decision Making Under Uncertainty by Sequential Sampling.

Neural computation·2026
Same journal

DROP: Distributional and Regular Optimism and Pessimism for Reinforcement Learning.

Neural computation·2026
Same journal

Hierarchical Active Inference Using Successor Representations.

Neural computation·2026
Same journal

W-Kernel and Its Principal Space for Frequentist Evaluation of Bayesian Estimators.

Neural computation·2026
Same journal

A Hidden Markov Model-Inspired Sequence Classification Method for Hyperdimensional Computing.

Neural computation·2026
Same journal

Sparse Graphical Modeling for Electrophysiological Phase-Based Connectivity Using Circular Statistics.

Neural computation·2026
See all related articles

This study introduces two novel boosting algorithms, Eta-Boost.M and normalized Eta-Boost.M, designed to enhance classification robustness against mislabeled data. Experiments confirm their effectiveness in handling noisy labels in multiclass classification tasks.

Area of Science:

  • Machine Learning
  • Computer Science

Background:

  • Classification problems often suffer from mislabeled data, impacting model performance.
  • Robustness against label noise is crucial for reliable machine learning models.

Discussion:

  • This research proposes two novel boosting algorithms, Eta-Boost.M and normalized Eta-Boost.M, leveraging Eta-divergence.
  • These algorithms are specifically designed to be robust against mislabeling in multiclass classification.
  • Theoretical analysis supports the robustness properties of the proposed boosting methods.

Key Insights:

  • The proposed Eta-Boost.M and normalized Eta-Boost.M algorithms demonstrate significant robustness against mislabeled data.
  • Empirical validation on synthetic and real datasets confirms the practical effectiveness of these methods.

Related Experiment Videos

  • The study validates the theoretical underpinnings of the algorithms in handling label noise.

    • Outlook:

    • Further research could explore the application of these algorithms to other types of data noise.
    • Investigating the scalability of Eta-Boost.M and normalized Eta-Boost.M for large-scale datasets is a potential future direction.
    • Exploring variations of Eta-divergence for enhanced robustness could be a promising avenue.