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

Survival Tree01:19

Survival Tree

146
Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
 Building a Survival Tree
Constructing a...
146
Reducing Line Loss01:18

Reducing Line Loss

190
In a three-phase circuit, line loss is an indicator of energy dissipated as heat due to the resistance of transmission lines. To address this, incorporating transformers into the system—a step-up transformer at the source and a step-down transformer at the load—is a strategic solution. Two three-phase transformers are introduced to improve this.
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss...
190
Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving01:29

Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving

100
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...
100

You might also read

Related Articles

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

Sort by
Same author

Statistical Extraction Method for Revealing Key Factors from Posture before Initiating Successful Throwing Technique in Judo.

Sensors (Basel, Switzerland)·2021
Same author

Stream-Based Visually Lossless Data Compression Applying Variable Bit-Length ADPCM Encoding.

Sensors (Basel, Switzerland)·2021
See all related articles

Related Experiment Video

Updated: Sep 1, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

488

Heuristic Method for Minimizing Model Size of CNN by Combining Multiple Pruning Techniques.

Danhe Tian1,2, Shinichi Yamagiwa3,4, Koichi Wada3

  • 1Doctoral Program in Computer Science, University of Tsukuba, 1-1-1 Tennodai, Tsukuba 305-8573, Ibaraki, Japan.

Sensors (Basel, Switzerland)
|August 12, 2022
PubMed
Summary

This study introduces a novel heuristic method combining structured and unstructured network pruning to significantly reduce deep learning model size while maintaining accuracy. The combined approach achieves higher compression ratios than individual methods, demonstrating superior efficiency.

Keywords:
convolutional neural networkdeep learningnetwork pruningobject detection

More Related Videos

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.3K
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.9K

Related Experiment Videos

Last Updated: Sep 1, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

488
Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.3K
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.9K

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Machine Learning

Background:

  • Deep learning models are computationally intensive, necessitating compression techniques.
  • Network pruning, categorized into structured and unstructured methods, aims to reduce model size by removing redundant components.
  • Existing methods have limitations, such as potential accuracy reduction (structured) or inadequate pruning (unstructured).

Purpose of the Study:

  • To propose a heuristic method for minimizing deep learning model size by combining structured and unstructured pruning.
  • To maintain target accuracy while achieving higher compression ratios.
  • To evaluate the effectiveness of the combined pruning approach against individual methods.

Main Methods:

  • Implemented a heuristic algorithm combining network slimming (structured pruning) and deep compression (unstructured pruning).
  • Evaluated the proposed method on state-of-the-art Convolutional Neural Network (CNN) models: VGGNet, ResNet, and DenseNet.
  • Tested performance using the CIFAR-10 dataset.

Main Results:

  • The combined pruning method achieved higher compression ratios compared to using structured or unstructured pruning individually.
  • Demonstrated superior performance and compression efficiency.
  • Achieved a 13x model size reduction and a 15x pruning time reduction for VGGNet compared to brute-force search.

Conclusions:

  • The proposed heuristic method effectively combines structured and unstructured pruning for significant model compression.
  • The combined approach outperforms individual pruning techniques in terms of compression ratio and efficiency.
  • This method offers a promising solution for compressing deep learning models without compromising accuracy.