Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jul 12, 2026

A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data

A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data

Published on: September 25, 2021

Feature-indistinguishable machine unlearning via negative-hot label encoding and class weight masking.

Jiali Wang¹, Hongxia Bie², Zhao Jing¹

¹Intelligent Media Computing Center, School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing, 100876, China.

Scientific Reports

|March 3, 2026

Summary

Related Concept Videos

Generalization, Discrimination, and Extinction

Generalization, Discrimination, and Extinction

Generalization, discrimination, and extinction are key concepts in operant conditioning that influence how behaviors are learned and maintained.
Generalization occurs when a behavior reinforced in one context is performed in similar situations. For instance, a student who studies diligently for calculus and receives excellent grades might apply the same study habits to psychology and history, expecting similar results. Generalization shows how learning in one setting can influence behavior in...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

A lightweight perceptual-guided VQVAE for high-fidelity image compression.

Scientific reports·2026

Same author

Siamese evolutionary masking: Enhancing the generalization of self-supervised medical image segmentation model.

Artificial intelligence in medicine·2026

Same author

Masked autoencoders with generalizable self-distillation for skin lesion segmentation.

Medical & biological engineering & computing·2024

Same author

NRF1 Alleviated Oxidative Stress of Glioblastoma Cells by Regulating NOR1.

Folia biologica·2023

Same author

Intestinal microbiome dysbiosis in alcohol-dependent patients and its effect on rat behaviors.

mBio·2023

Same author

Complete mitochondrial genome and its phylogenetic implications of Rhinogobio nasutus, an endemic species from the Yellow River.

Molecular biology reports·2023

This study introduces a new machine unlearning method using Negative-Hot Label Encoding (NHLE) for efficient data privacy. The technique selectively forgets data classes with minimal impact on overall model performance.

Area of Science:

Artificial Intelligence
Machine Learning
Deep Learning

Background:

Data privacy and regulatory compliance are increasingly critical in deep learning.
Existing machine unlearning methods face challenges with data access, computational cost, and performance degradation on retained data.

Purpose of the Study:

To develop an efficient and selective machine unlearning framework.
To address limitations of current unlearning approaches by minimizing computational overhead and preserving performance on non-forgotten data.

Main Methods:

Proposed a novel unlearning framework integrating label encoding fine-tuning and class weight masking.
Introduced Negative-Hot Label Encoding (NHLE) to suppress target class discriminability.
Utilized a small number of samples from forgotten classes for iterative fine-tuning.

Keywords:

Feature indistinguishable representation Label encoding Machine unlearning

Related Experiment Videos

Last Updated: Jul 12, 2026

A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data

A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data

Published on: September 25, 2021

Main Results:

Achieved near-zero classification accuracy on forgotten data across multiple visual datasets.
Demonstrated a minimal reduction in accuracy (≤0.035) on retained data.
Showcased the efficiency and selectivity of the proposed unlearning framework.

Conclusions:

The NHLE-integrated framework offers an effective solution for selective machine unlearning.
This approach enhances data privacy and regulatory compliance in deep learning models.
The method balances efficient forgetting with the preservation of model utility on remaining data.