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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Through the Looking Glass: A Dual Perspective on Weakly Supervised Few-Shot Segmentation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same author

Location Matters: Frequency-Spatial Dual-Space Adaptation for Cross-Domain Few-Shot Segmentation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same author

Impact of adjuvant breast radiotherapy on the risk and the survival of second primary lung cancer: a large population-based study.

Japanese journal of clinical oncology·2026
Same author

Topology-Preserving Deep Hashing for Ultrafast Drone-Dominated Object Detection.

IEEE transactions on neural networks and learning systems·2026
Same author

Hypomagnetic Field Enhances U2OS Cell Proliferation and Migration by Promoting β-Catenin Phosphorylation and Upregulating FN1 and LOX Expression.

Cells·2026
Same author

High-throughput screening of EGFR/Ca<sup>2+</sup> signaling modulators in cardiac hypertrophy using a tetrahedral DNA nanostructure-based hESC platform.

Journal of pharmaceutical analysis·2026

Related Experiment Video

Updated: Aug 28, 2025

Mass Spectrometry-Guided Genome Mining as a Tool to Uncover Novel Natural Products
11:13

Mass Spectrometry-Guided Genome Mining as a Tool to Uncover Novel Natural Products

Published on: March 12, 2020

11.0K

Deep Metric Learning Based on Meta-Mining Strategy With Semiglobal Information.

Xiruo Jiang, Sheng Liu, Xili Dai

    IEEE Transactions on Neural Networks and Learning Systems
    |September 15, 2022
    PubMed
    Summary

    This study introduces a meta-mining strategy with semiglobal information (MMSI) for deep metric learning (DML). MMSI enhances sample weighting using meta-learning and a large memory dictionary, achieving state-of-the-art performance.

    More Related Videos

    Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
    07:15

    Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

    Published on: August 16, 2020

    6.9K
    Heuristic Mining of Hierarchical Genotypes and Accessory Genome Loci in Bacterial Populations
    08:03

    Heuristic Mining of Hierarchical Genotypes and Accessory Genome Loci in Bacterial Populations

    Published on: December 7, 2021

    2.3K

    Related Experiment Videos

    Last Updated: Aug 28, 2025

    Mass Spectrometry-Guided Genome Mining as a Tool to Uncover Novel Natural Products
    11:13

    Mass Spectrometry-Guided Genome Mining as a Tool to Uncover Novel Natural Products

    Published on: March 12, 2020

    11.0K
    Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
    07:15

    Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

    Published on: August 16, 2020

    6.9K
    Heuristic Mining of Hierarchical Genotypes and Accessory Genome Loci in Bacterial Populations
    08:03

    Heuristic Mining of Hierarchical Genotypes and Accessory Genome Loci in Bacterial Populations

    Published on: December 7, 2021

    2.3K

    Area of Science:

    • Computer Science
    • Artificial Intelligence
    • Machine Learning

    Background:

    • Deep metric learning (DML) shows significant success.
    • Existing adaptive sample mining strategies in DML are limited by small memory, typically one training batch.

    Purpose of the Study:

    • To introduce a novel data-driven method, meta-mining strategy with semiglobal information (MMSI), for deep metric learning.
    • To overcome the memory limitations of existing DML methods by employing meta-learning for sample weighting throughout the entire training process.

    Main Methods:

    • Developed MMSI, a meta-learning approach to learn sample weights adaptively.
    • Integrated validation set information and a large dictionary (memory) to provide richer training data context.
    • Proposed a unified theoretical framework for pairwise and tripletwise metric learning losses.

    Main Results:

    • MMSI demonstrated promising performance by leveraging richer information from the validation set and dictionary.
    • The method achieved state-of-the-art or highly competitive results on CUB200-2011, Cars-196, and Stanford Online Products datasets.
    • The generalized theoretical framework facilitates the application of MMSI to various existing DML methods.

    Conclusions:

    • MMSI effectively enhances deep metric learning by utilizing meta-learning and extensive data information.
    • The proposed unified framework offers new theoretical insights and broad applicability for DML.
    • The method represents a significant advancement in adaptive sample mining for DML.