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

Protein Networks02:26

Protein Networks

4.2K
An organism can have thousands of different proteins, and these proteins must cooperate to ensure the health of an organism. Proteins bind to other proteins and form complexes to carry out their functions. Many proteins interact with multiple other proteins creating a complex network of protein interactions.
These interactions can be represented through maps depicting protein-protein interaction networks, represented as nodes and edges. Nodes are circles that are representative of a protein,...
4.2K
Neuroplasticity01:01

Neuroplasticity

926
Neuroplasticity reflects the brain's remarkable capacity to adapt and evolve, responding dynamically to learning, experiences, or injury by reorganizing its neural circuitry. This reorganization involves creating new neural connections and refining old ones through a series of biological processes that contribute to the brain's lifelong development and adaptability.
926
Neural Circuits01:25

Neural Circuits

1.9K
Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
1.9K
Reducing Line Loss01:18

Reducing Line Loss

218
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...
218
Survival Tree01:19

Survival Tree

182
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...
182
Neural Regulation01:37

Neural Regulation

40.5K
Digestion begins with a cephalic phase that prepares the digestive system to receive food. When our brain processes visual or olfactory information about food, it triggers impulses in the cranial nerves innervating the salivary glands and stomach to prepare for food.
40.5K

You might also read

Related Articles

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

Sort by
Same author

Paving the Way for Point Cloud Video Representation Learning Using a PDE Model.

IEEE transactions on pattern analysis and machine intelligence·2026
Same author

A TSST-1 structural motif disrupts endothelial programs required for vascular regeneration.

bioRxiv : the preprint server for biology·2026
Same author

Synthesis of Carborane-Containing Esters of Natural l-Amino Acids via Construction of a Carborane Cage Using Alkyne-Insertion Reactions.

Inorganic chemistry·2026
Same author

ClC-3 mediates angiotensin II-induced endothelial dysfunction by inhibiting Akt-Hsp90-eNOS signaling pathway.

Frontiers in pharmacology·2026
Same author

EvaNet: Toward More Efficient and Consistent Infrared and Visible Image Fusion Assessment.

IEEE transactions on pattern analysis and machine intelligence·2026
Same author

ScRNA-Seq Deciphers an Autocrine EFNA1-EPHA1 Loop That Reprograms the Microenvironment in Hepatocellular Carcinoma.

Journal of hepatocellular carcinoma·2026

Related Experiment Video

Updated: Oct 16, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

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

Published on: December 15, 2023

690

Weak sub-network pruning for strong and efficient neural networks.

Qingbei Guo1, Xiao-Jun Wu2, Josef Kittler3

  • 1Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence, Jiangnan University, Wuxi 214122, China; Shandong Provincial Key Laboratory of Network based Intelligent Computing, University of Jinan, Jinan 250022, China.

Neural Networks : the Official Journal of the International Neural Network Society
|October 15, 2021
PubMed
Summary

This study introduces Weak Sub-network Pruning (WSP), a novel method to compress deep convolutional neural networks (CNNs). WSP efficiently prunes entire networks in two stages, enabling deployment on resource-constrained devices with minimal performance loss.

Keywords:
AccelerationCompressionDeep neural networkWeak sub-network pruning

More Related Videos

Modeling the Functional Network for Spatial Navigation in the Human Brain
05:55

Modeling the Functional Network for Spatial Navigation in the Human Brain

Published on: October 13, 2023

1.2K

Related Experiment Videos

Last Updated: Oct 16, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

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

Published on: December 15, 2023

690
Modeling the Functional Network for Spatial Navigation in the Human Brain
05:55

Modeling the Functional Network for Spatial Navigation in the Human Brain

Published on: October 13, 2023

1.2K

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Machine Learning

Background:

  • Deep convolutional neural networks (CNNs) are computationally intensive, limiting their deployment on resource-constrained hardware.
  • Existing pruning methods often focus on small granularities (weights, kernels, filters), requiring iterative pruning for high compression ratios and risking performance degradation.

Purpose of the Study:

  • To develop an efficient and effective pruning method for CNNs that addresses the limitations of existing approaches.
  • To enable the deployment of compressed CNNs on resource-constrained devices without significant loss in accuracy.

Main Methods:

  • The proposed Weak Sub-network Pruning (WSP) method theoretically analyzes the relationship between activation/gradient sparsity and channel saliency.
  • WSP prunes the weakest sub-network in a one-shot manner by identifying and removing channels with minimal contribution to feed-forward and feed-backward processes.
  • The process involves two non-iterative stages: pruning the weakest sub-network and globally fine-tuning the resulting strong sub-network.

Main Results:

  • WSP achieves high compression ratios and acceleration for deep and wide CNN architectures, including VGG16 and ResNet-50.
  • The method demonstrates superior performance in classification, domain adaptation, and object detection tasks across various benchmarks (ImageNet-1K, CIFAR-10, CUB-200, PASCAL VOC).
  • WSP effectively compresses both convolutional and fully-connected layers, leading to simultaneous compression and acceleration.

Conclusions:

  • Weak Sub-network Pruning (WSP) offers an effective strategy for compressing deep convolutional neural networks (CNNs).
  • The method facilitates the deployment of efficient CNNs on resource-constrained hardware while maintaining high accuracy.
  • WSP provides a novel, non-iterative approach to network compression with broad applicability in computer vision tasks.