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

Related Concept Videos

Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving

Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving

Mechanistic models play a crucial role in algorithms for numerical problem-solving, particularly in nonlinear mixed effects modeling (NMEM). These models aim to minimize specific objective functions by evaluating various parameter estimates, leading to the development of systematic algorithms. In some cases, linearization techniques approximate the model using linear equations.
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...

You might also read

Related Articles

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

Sort by

Same author

Correction: Accelerating automatic model finding with layer replications case study of MobileNetV2.

PloS one·2025

Same author

Parallel simulation of HGMS of weakly magnetic nanoparticles in irrotational flow of inviscid fluid.

TheScientificWorldJournal·2014

See all related articles

Search research articles

Related Experiment Video

Updated: Jun 15, 2025

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

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

Published on: August 16, 2020

Accelerating automatic model finding with layer replications case study of MobileNetV2.

Kritpawit Soongswang¹, Chantana Chantrapornchai¹

¹Department of Computer Engineering, Kasetsart University, Bangkok, Thailand.

|August 22, 2024

Summary

This study introduces an efficient method for optimizing 3D face recognition models by intelligently replicating layers, significantly reducing search time and improving accuracy. The approach enhances MobileNetV2 performance using distributed and concurrent training strategies.

More Related Videos

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning

Published on: May 10, 2024

Related Experiment Videos

Last Updated: Jun 15, 2025

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

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

Published on: August 16, 2020

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning

Published on: May 10, 2024

Area of Science:

Computer Vision
Machine Learning
Deep Learning Architectures

Background:

3D face recognition models require extensive computational resources for architecture searching.
Optimizing model architecture is crucial for enhancing accuracy and reducing training time.
MobileNetV2 is a popular architecture, but its application in 3D face recognition can be further improved.

Purpose of the Study:

To develop a method for reducing model architecture searching time in 3D face recognition.
To enhance the accuracy of MobileNetV2 for 3D face recognition tasks through layer replication.
To investigate the effectiveness of distributed data-parallel and concurrent training for accelerating the model finding process.

Main Methods:

Proposed an algorithm to identify optimal layer replication configurations for neural networks.
Utilized MobileNetV2 as a case study for 3D face recognition.
Implemented and compared distributed data-parallel training and concurrent model training acceleration methods.
Evaluated the automatic model finding process for layer replication under various conditions.

Main Results:

Achieved accuracy improvements of up to 6% over previous 3D MobileNetV2 work and 8% over vanilla MobileNetV2.
Reduced model training time by up to 75% using distributed data-parallel training across four GPUs compared to single-GPU training.
The concurrent training approach for automatic model finding was 1,932 minutes faster than distributed training.

Conclusions:

The proposed automatic model finding process for layer replication is effective in optimizing 3D face recognition models.
Layer replication, combined with efficient training strategies, significantly enhances model accuracy and reduces computational overhead.
This method offers a practical solution for accelerating the development of high-performance 3D face recognition systems.