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: Dec 29, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

5.1K

A Privacy-Preserving Multi-Task Learning Framework for Face Detection, Landmark Localization, Pose Estimation, and

Chen Zhang1, Xiongwei Hu1, Yu Xie1

  • 1School of Computer Science and Technology, Xidian University, Xi'an, China.

Frontiers in Neurorobotics
|January 30, 2020
PubMed
Summary

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

[Etiology and pathogens of fungal endophthalmitis].

[Zhonghua yan ke za zhi] Chinese journal of ophthalmology·2015
Same author

Eucommia ulmoides Oliv. bark aqueous extract inhibits osteoarthritis in a rat model of osteoarthritis.

Journal of ethnopharmacology·2015
Same author

Aucubin prevents interleukin-1 beta induced inflammation and cartilage matrix degradation via inhibition of NF-κB signaling pathway in rat articular chondrocytes.

International immunopharmacology·2015
Same author

Treatment with recombinant lubricin attenuates osteoarthritis by positive feedback loop between articular cartilage and subchondral bone in ovariectomized rats.

Bone·2015
Same author

Tet1-mediated DNA demethylation regulates neuronal cell death induced by oxidative stress.

Scientific reports·2015
Same author

Authors' reply.

Arthroscopy : the journal of arthroscopic & related surgery : official publication of the Arthroscopy Association of North America and the International Arthroscopy Association·2015
This summary is machine-generated.

This study introduces a privacy-preserving multi-task learning (MTL) method using differential privacy to protect sensitive face data. The approach ensures data security without compromising the accuracy of face processing tasks.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Data Privacy

Background:

  • Multi-task learning (MTL) is widely used for face processing tasks like detection and recognition.
  • Training data for face processing can contain sensitive information, posing privacy risks.
  • Existing MTL methods may not adequately protect private data from model analysis.

Purpose of the Study:

  • To develop a novel privacy-preserving multi-task learning approach for face processing.
  • To safeguard sensitive information within raw face datasets.
  • To enhance learning efficiency and prediction accuracy while ensuring privacy.

Main Methods:

  • Utilized differential private stochastic gradient descent (DPSGD) for optimizing the MTL model.
  • Incorporated calibrated noise addition to loss function gradients for data privacy.
Keywords:
balance different learning tasksdifferential privacy guaranteesdifferential private stochastic gradient descentmulti-task learningprivacy preserving

Related Experiment Videos

Last Updated: Dec 29, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

5.1K
  • Employed homoscedastic uncertainty to balance and weigh multiple learning tasks.

    • Main Results:

    • The proposed method provides differential privacy guarantees for training data.
    • Achieved comparable or improved accuracy in face processing tasks compared to non-private methods.
    • Demonstrated effective balancing of multiple tasks under a privacy budget.

    Conclusions:

    • The novel privacy-preserving MTL approach effectively protects sensitive face data.
    • DPSGD and uncertainty weighting offer a viable solution for secure and accurate face analysis.
    • The method maintains high performance while adhering to privacy requirements.