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

Hedgehog Signaling Pathway

Hedgehog Signaling Pathway

The Hedgehog gene (Hh) was first discovered due to its control of the growth of disorganized, hair-like bristles phenotype in Drosophila, much like hedgehog spines. Hh plays a crucial role in the development of organs and the maintenance of homeostasis in both invertebrates and vertebrates. However, while Drosophila has only one Hh protein, mammals have multiple functional Hedgehog proteins - Sonic (Shh), Desert (Dhh), and Indian Hedgehog (Ihh). All of these homologous proteins have adapted to...

You might also read

Related Articles

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

Sort by

Same author

Evaluation of nucleic acid-based matrix-assisted laser desorption/ionization time-of-flight mass spectrometry for rapid identification of <i>Mycobacterium tuberculosis</i> and drug resistance in patients with retreatment tuberculosis.

Frontiers in microbiology·2026

Same author

Cold-Driven Thermoelectric Patch for Postoperative Tumor Control.

ACS nano·2026

Same author

A QuantiFERON<sup>®</sup>-TB Gold Plus-Based Model for Predicting Latent Tuberculosis Infection in Patients with Diabetes.

Infection and drug resistance·2026

Same author

Dissolved Organic Matter Chemistry Modulates Biochar Effects on Phenanthrene Biodegradation in Agricultural Soil: Bioavailability and Microbial Responses.

Journal of agricultural and food chemistry·2026

Same author

Identification of serum biomarkers in acute aortic dissection using tissue-informed metabolomics methods.

Scientific reports·2026

Same author

Integrated Gut-Brain Axis Response to Freezing and Recovery in Freeze-Tolerant Fish, <i>Perccottus glenii</i>.

Animals : an open access journal from MDPI·2026

Same journal

Granular Ball-Based Noise-Resistant Fuzzy Multineighborhood Feature Selection via Label Enhancement and Feature Graph.

IEEE transactions on neural networks and learning systems·2026

Same journal

Fighting Evolving Spam With ARTMAP Models: A Noise-Resilient Online Detection Framework.

IEEE transactions on neural networks and learning systems·2026

Same journal

HyperSAT: Unsupervised Hypergraph Neural Networks for Weighted MaxSAT Problems.

IEEE transactions on neural networks and learning systems·2026

Same journal

Negation of Basic Belief Assignment in Multisource Information Fusion on Dempster-Shafer Theory With Applications in Pattern Classification.

IEEE transactions on neural networks and learning systems·2026

Same journal

Intervention Feasible Region and Driver Risk Capacity Aware Human-Machine Collaborative Safe Trajectory Planning.

IEEE transactions on neural networks and learning systems·2026

Same journal

A Unified Differential Denoising Learning Framework With a Pre-Trained Model and Fuzzy Graph Networks for Drug-Drug Interaction Prediction.

IEEE transactions on neural networks and learning systems·2026

See all related articles

Search research articles

Related Experiment Video

Updated: Jul 9, 2025

In situ Protocol for Butterfly Pupal Wings Using Riboprobes

In situ Protocol for Butterfly Pupal Wings Using Riboprobes

Published on: May 28, 2007

Lite It Fly: An All-Deformable-Butterfly Network.

Rui Lin, Jason Chun Lok Li, Jiajun Zhou

IEEE Transactions on Neural Networks and Learning Systems

|November 28, 2023

Summary

This summary is machine-generated.

This study introduces Deformable Butterfly (DeBut) layers for deep neural network (DNN) compression. DeBut layers achieve extreme network sparsity and compression, significantly outperforming existing methods.

More Related Videos

Fully Automated Leg Tracking in Freely Moving Insects using Feature Learning Leg Segmentation and Tracking FLLIT

Fully Automated Leg Tracking in Freely Moving Insects using Feature Learning Leg Segmentation and Tracking FLLIT

Published on: April 23, 2020

Two Algorithms for High-throughput and Multi-parametric Quantification of Drosophila Neuromuscular Junction Morphology

Two Algorithms for High-throughput and Multi-parametric Quantification of Drosophila Neuromuscular Junction Morphology

Published on: May 3, 2017

Related Experiment Videos

Last Updated: Jul 9, 2025

In situ Protocol for Butterfly Pupal Wings Using Riboprobes

In situ Protocol for Butterfly Pupal Wings Using Riboprobes

Published on: May 28, 2007

Fully Automated Leg Tracking in Freely Moving Insects using Feature Learning Leg Segmentation and Tracking FLLIT

Fully Automated Leg Tracking in Freely Moving Insects using Feature Learning Leg Segmentation and Tracking FLLIT

Published on: April 23, 2020

Two Algorithms for High-throughput and Multi-parametric Quantification of Drosophila Neuromuscular Junction Morphology

Two Algorithms for High-throughput and Multi-parametric Quantification of Drosophila Neuromuscular Junction Morphology

Published on: May 3, 2017

Area of Science:

Computer Science
Artificial Intelligence
Machine Learning

Background:

Deep neural networks (DNNs) commonly use convolutional and fully connected layers.
Network compression is crucial for efficient deployment, with traditional methods including pruning and low-rank decomposition.

Purpose of the Study:

To introduce and analyze Deformable Butterfly (DeBut) layers as a novel approach to DNN compression.
To demonstrate the effectiveness of homogenizing DNNs into exclusively DeBut layers for extreme sparsity and compression.

Main Methods:

Decomposing filter matrices into generalized, butterfly-like factors using DeBut.
Developing an automated DeBut chain generator to create fully DeBut networks.
Evaluating network performance through various examples and hardware benchmarks.

Main Results:

An intimate link between DeBut and depthwise/pointwise convolutions was revealed, explaining DeBut's performance.
Achieved extreme sparsity and compression by homogenizing DNNs into all-DeBut layers.
Compressed a PointNet to less than 5% of its original parameters with minimal accuracy loss (<5%).

Conclusions:

DeBut layers offer a compression strategy orthogonal to traditional methods.
All-DeBut networks demonstrate significant advantages in sparsity and parameter reduction.
This approach sets a new record for DNN compression, particularly for PointNets.