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

Neuroplasticity01:01

Neuroplasticity

314
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.
314
Association Areas of the Cortex01:21

Association Areas of the Cortex

5.2K
Association areas are regions of the cerebral cortex that do not have a specific sensory or motor function. Instead, they integrate and interpret information from various sources to enable higher cognitive processes such as memory, learning, and decision-making. Some key association areas include the following:
Prefrontal Association Area: This area is located in the frontal lobe and is involved in planning, decision-making, and moderating social behavior. It connects with primary motor areas,...
5.2K

You might also read

Related Articles

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

Sort by
Same author

Comparison of frequency-resolved optical polarization gating induced by molecular alignment and Kerr effects.

Optics letters·2012
Same author

Direct transformation of simple enals to 3,4-disubstituted benzaldehydes under mild reaction conditions via an organocatalytic regio- and chemoselective dimerization cascade.

Chemistry (Weinheim an der Bergstrasse, Germany)·2012
Same author

[Digital anatomy of the perforator flap in the thigh].

Zhonghua zheng xing wai ke za zhi = Zhonghua zhengxing waike zazhi = Chinese journal of plastic surgery·2012
Same author

[Value of methylation-specific mutiplex ligation-dependent probe in the diagnosis of Prader-Willi syndrome].

Zhongguo dang dai er ke za zhi = Chinese journal of contemporary pediatrics·2012
Same author

Elevated local TGF-β1 level predisposes a closed bone fracture to tuberculosis infection.

Medical hypotheses·2012
Same author

Modulation of P-glycoprotein expression by triptolide in adriamycin-resistant K562/A02 cells.

Oncology letters·2012

Related Experiment Video

Updated: Jun 14, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

379

SAAF: Self-Adaptive Attention Factor-Based Taylor-Pruning on Convolutional Neural Networks.

Yiheng Lu, Maoguo Gong, Kaiyuan Feng

    IEEE Transactions on Neural Networks and Learning Systems
    |August 30, 2024
    PubMed
    Summary
    This summary is machine-generated.

    This study introduces a self-adaptive attention factor (SAAF) to enhance Taylor-based pruning in convolutional neural networks (CNNs). SAAF improves accuracy in pruned models, especially at high pruning rates, by intelligently recovering filters.

    More Related Videos

    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

    492
    Measuring Attention and Visual Processing Speed by Model-based Analysis of Temporal-order Judgments
    13:00

    Measuring Attention and Visual Processing Speed by Model-based Analysis of Temporal-order Judgments

    Published on: January 23, 2017

    9.9K

    Related Experiment Videos

    Last Updated: Jun 14, 2025

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

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    379
    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

    492
    Measuring Attention and Visual Processing Speed by Model-based Analysis of Temporal-order Judgments
    13:00

    Measuring Attention and Visual Processing Speed by Model-based Analysis of Temporal-order Judgments

    Published on: January 23, 2017

    9.9K

    Area of Science:

    • Computer Science
    • Artificial Intelligence
    • Machine Learning

    Background:

    • Convolutional Neural Networks (CNNs) require significant computational resources.
    • Pruning techniques aim to reduce model size and computational cost.
    • Taylor-based pruning is efficient but suffers accuracy loss at high pruning rates.

    Purpose of the Study:

    • To improve the accuracy of CNN models pruned using the Taylor-based method, particularly at high pruning rates.
    • To introduce a novel self-adaptive attention factor (SAAF) to mitigate accuracy degradation.
    • To maintain model compression benefits while enhancing performance.

    Main Methods:

    • Proposed a self-adaptive attention factor (SAAF) integrated with Taylor-based pruning.
    • SAAF leverages the remaining filter ratio during early pruning stages.
    • Implemented filter recovery based on SAAF to protect critical filters.

    Main Results:

    • The proposed SAAF method significantly improved accuracy compared to traditional Taylor-based pruning on various datasets (CIFAR-10, Tiny-ImageNet, ImageNet-1000).
    • Minimal sacrifice in parameter reduction and FLOPs reduction was observed.
    • Outperformed other pruning algorithms in terms of accuracy-computation trade-off.

    Conclusions:

    • SAAF effectively addresses the accuracy degeneration issue of Taylor-based pruning at high pruning rates.
    • The method offers a robust approach for compressing CNNs while preserving or enhancing performance.
    • SAAF demonstrates broad applicability across different datasets and network architectures (VGG-16, ResNet-50).