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 Video

Updated: Jul 17, 2026

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

Discriminative transfer feature learning for unsupervised domain adaptation.

Mengyao Li1, Zhengming Li2, Jinghua Jiang3

  • 1School of Cyber Security, Guangdong Polytechnic Normal University, Guangzhou, 510630, China.

Neural Networks : the Official Journal of the International Neural Network Society
|July 15, 2026
PubMed
Summary

Related Concept Videos

Generalization, Discrimination, and Extinction01:24

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

Multistimuli-Controlled Topological Nucleation of Skyrmion Loops and Monopoles in Liquid Crystals.

Physical review letters·2026
Same author

Dynamic creation of topological solitons via nematic vortex lines.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same author

Light-Driven Undulatory Locomotion of a Liquid Crystal Gel Robot for Programmable 3D Navigation.

ACS applied materials & interfaces·2026
Same author

Glycochenodeoxycholic acid promotes hepatocarcinogenesis by inducing hepatic progenitor cell differentiation into cancer-associated fibroblasts via sphingosine-1-phosphate receptor 2 signalling.

Biomarker research·2025
Same author

Programmable double traveling waves in living liquid crystals.

Nature communications·2025
Same author

Transmembrane BAX inhibitor motif containing 1 inhibition of lysosomal degradation of TGF-β receptor 1 suppresses cellular senescence and hepatocarcinogenesis.

The Journal of biological chemistry·2025

This study introduces discriminative transfer feature learning (DTFL) for unsupervised domain adaptation. DTFL enhances classification by preserving data locality and refining pseudo-labels for improved feature discriminability.

Area of Science:

  • Machine Learning
  • Computer Vision
  • Data Science

Background:

  • Unsupervised domain adaptation (UDA) methods often focus on domain-invariant and category-discriminative features.
  • The role of ordinal data locality in constructing discriminative terms for UDA remains underexplored.

Purpose of the Study:

  • To propose a novel UDA method, Discriminative Transfer Feature Learning (DTFL), for pattern classification.
  • To enhance the class-wise discriminability of transferable features by incorporating ordinal locality preservation.

Main Methods:

  • DTFL constructs a class-aware ordinal locality-preserving term (COLP) to maintain neighborhood relationships and geometric structure.
  • Label consistency within domains is employed to refine target pseudo-labels.
  • Iterative mutual benefit between DTFL and refined labels enhances performance.
Keywords:
Discriminative transfer feature learningOrdinal locality preservingPattern classificationPseudo-labelUnsupervised domain adaptation

Related Experiment Videos

Last Updated: Jul 17, 2026

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

Main Results:

  • Experiments on visual and intrusion detection datasets show DTFL outperforms existing UDA methods.
  • The proposed COLP term effectively enhances feature discriminability and preserves geometric structures.
  • Iterative refinement of pseudo-labels improves classification accuracy.

Conclusions:

  • DTFL offers a significant advancement in UDA by integrating ordinal locality preservation.
  • The method demonstrates superior classification performance across diverse datasets.
  • Future work could explore further applications of locality-preserving techniques in UDA.