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

178
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...
178

You might also read

Related Articles

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

Sort by
Same author

Diagnosis of superficial ailments using infrared thermal imaging and CapsNet.

Journal of thermal biology·2025
Same author

An information-rich sampling technique over spatio-temporal CNN for classification of human actions in videos.

Multimedia tools and applications·2022
Same author

AutoTune: Automatically Tuning Convolutional Neural Networks for Improved Transfer Learning.

Neural networks : the official journal of the International Neural Network Society·2020
Same author

diffGrad: An Optimization Method for Convolutional Neural Networks.

IEEE transactions on neural networks and learning systems·2019
Same author

Multiple Pyramids Based Image Inpainting Using Local Patch Statistics and Steering Kernel Feature.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2019
Same author

A Group-Based Image Inpainting Using Patch Refinement in MRF Framework.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2017
Same journal

Dynamic analysis and reliable mechanical optimization application of ring HNN effected with a memristive neuron.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

DAFF-Net: A detection and search method for small-scale low surface brightness galaxies.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Quasi-synchronization for complex networks with hybrid pinning intermittent control.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Physics-encoded convolutional neural operators for parametric PDEs: A convergence-guaranteed framework via pre-computed kernel fields.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Exploiting audio-visual modalities in videos: Object detection via multi-stage bilateral coupling network.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Reliability-aware modality completion with cross-modal distillation for federated learning with missing modalities.

Neural networks : the official journal of the International Neural Network Society·2026
See all related articles

Related Experiment Video

Updated: Oct 6, 2025

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations
12:27

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

Published on: February 15, 2017

7.1K

HRel: Filter pruning based on High Relevance between activation maps and class labels.

C H Sarvani1, Mrinmoy Ghorai1, Shiv Ram Dubey2

  • 1Computer Vision Group, Indian Institute of Information Technology, Sri City, Chittoor, Andhra Pradesh 517646, India.

Neural Networks : the Official Journal of the International Neural Network Society
|January 18, 2022
PubMed
Summary
This summary is machine-generated.

This study introduces a filter pruning method using Mutual Information (MI) to identify and remove less important filters in deep learning models. The High Relevance (HRel) approach achieves state-of-the-art results, significantly reducing model size with minimal accuracy loss.

Keywords:
Activation mapsConvolutional Neural NetworksEntropyFilter pruningInformation Bottleneck theoryMutual information

More Related Videos

Classification of Neural Stem Cell Activation State In Vitro using Autofluorescence
06:56

Classification of Neural Stem Cell Activation State In Vitro using Autofluorescence

Published on: April 12, 2024

716
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 6, 2025

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations
12:27

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

Published on: February 15, 2017

7.1K
Classification of Neural Stem Cell Activation State In Vitro using Autofluorescence
06:56

Classification of Neural Stem Cell Activation State In Vitro using Autofluorescence

Published on: April 12, 2024

716
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, Deep Learning, Computer Vision

Background:

  • Deep neural networks, particularly Convolutional Neural Networks (CNNs), are computationally intensive, necessitating efficient model compression techniques.
  • Filter pruning is a crucial model compression strategy aimed at reducing the size and computational cost of CNNs.
  • Existing methods often lack a robust theoretical foundation for determining filter importance.

Purpose of the Study:

  • To propose a novel filter pruning method based on Information Bottleneck theory and Mutual Information (MI).
  • To enhance model compression by identifying and pruning filters with low relevance to class labels.
  • To demonstrate the efficacy of the proposed method across various CNN architectures and datasets.

Main Methods:

  • Utilized Information Bottleneck theory to define filter importance based on Mutual Information (MI) between filter activations and class labels.
  • Introduced the High Relevance (HRel) metric, quantifying the relationship between filter activation maps and annotations.
  • Applied the HRel pruning method to LeNet-5, VGG-16, ResNet-56, ResNet-110, and ResNet-50 architectures on MNIST, CIFAR-10, and ImageNet datasets.

Main Results:

  • Achieved state-of-the-art pruning results, significantly reducing Floating Point Operations (FLOPs) across all tested architectures.
  • Demonstrated substantial parameter reduction (e.g., 94.98% for VGG-16) with minimal impact on accuracy (e.g., 0.36% drop for VGG-16 top-1 accuracy).
  • Showcased drastic pruning of filters in LeNet-5 (from 20, 50 to 2, 3) with only a 0.52% accuracy decrease.

Conclusions:

  • The proposed HRel filter pruning method effectively reduces model complexity while preserving high accuracy, outperforming existing state-of-the-art techniques.
  • The method's reliance on activation map-class label relationships provides a theoretically grounded and effective approach to filter pruning.
  • Analysis of Information Plane dynamics confirmed the method's effectiveness and provided insights into the impact of pruning on CNNs.