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

1.4K
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.
1.4K
Neural Circuits01:25

Neural Circuits

2.5K
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...
2.5K
Neuron Structure01:31

Neuron Structure

230.2K
Overview
230.2K
Neuron Structure01:30

Neuron Structure

17.5K
Neurons are the main type of cell in the nervous system that generate and transmit electrochemical signals. They primarily communicate with each other using neurotransmitters at specific junctions called synapses. Neurons come in many shapes that often relate to their function, but most share three main structures: an axon and dendrites that extend out from a cell body.
Structure and Function of Neurons
The neuronal cell body—the soma— houses the nucleus and organelles vital to...
17.5K
The Role of Ion Channels in Neuronal Computation01:19

The Role of Ion Channels in Neuronal Computation

3.6K
A postsynaptic neuron usually receives numerous impulses from several other presynaptic neurons. The axon hillock of the postsynaptic neuron integrates all these signals and determines the likelihood of firing an action potential.
Sometimes a single EPSP is strong enough to induce an action potential in the postsynaptic neuron. However, multiple presynaptic inputs must often create EPSPs around the same time for the postsynaptic neuron to be sufficiently depolarized to fire an action potential....
3.6K

You might also read

Related Articles

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

Sort by
Same author

KnitLoRA: bridging low-rank adaptation as interwoven layers for deeper semantic reasoning.

Scientific reports·2026
Same author

Enhancing progressive ensemble learning via normalized extra-Gradient initialization.

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

TimeCNN: Refining inscross-variable interaction on time point for time series forecasting.

Neural networks : the official journal of the International Neural Network Society·2025
Same author

InstructLayout: Instruction-Driven 2D and 3D Layout Synthesis With Semantic Graph Prior.

IEEE transactions on pattern analysis and machine intelligence·2025
Same author

StoCFL: A stochastically clustered federated learning framework for Non-IID data with dynamic client participation.

Neural networks : the official journal of the International Neural Network Society·2025
Same author

Advocating for the Silent: Enhancing Federated Generalization for Nonparticipating Clients.

IEEE transactions on neural networks and learning systems·2025

Related Experiment Video

Updated: Jan 1, 2026

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

7.3K

Structured pruning of recurrent neural networks through neuron selection.

Liangjian Wen1, Xuanyang Zhang1, Haoli Bai2

  • 1SMILE Lab, School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu 610031, China.

Neural Networks : the Official Journal of the International Neural Network Society
|December 20, 2019
PubMed
Summary

This study introduces a structured pruning method for Recurrent Neural Networks (RNNs) to enable deployment on edge devices. The novel approach achieves significant speedups without performance loss, making complex AI models more accessible.

Keywords:
Feature selectionLearning sparse modelsModel compressionRecurrent neural networks

More Related Videos

Author Spotlight: Exploring Glial Influence in Experience-Dependent Synaptic Pruning During Critical Periods
07:13

Author Spotlight: Exploring Glial Influence in Experience-Dependent Synaptic Pruning During Critical Periods

Published on: March 1, 2024

985
Subtype-selective Electroporation of Cortical Interneurons
06:42

Subtype-selective Electroporation of Cortical Interneurons

Published on: August 18, 2014

9.1K

Related Experiment Videos

Last Updated: Jan 1, 2026

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

7.3K
Author Spotlight: Exploring Glial Influence in Experience-Dependent Synaptic Pruning During Critical Periods
07:13

Author Spotlight: Exploring Glial Influence in Experience-Dependent Synaptic Pruning During Critical Periods

Published on: March 1, 2024

985
Subtype-selective Electroporation of Cortical Interneurons
06:42

Subtype-selective Electroporation of Cortical Interneurons

Published on: August 18, 2014

9.1K

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Deep Learning

Background:

  • Recurrent Neural Networks (RNNs) offer significant capabilities but face deployment challenges on edge devices due to large sizes and computational demands.
  • Existing network pruning techniques, while reducing costs, often yield irregular sparse patterns that hinder practical speedups.

Purpose of the Study:

  • To develop a structured pruning method for RNNs that addresses the limitations of current techniques.
  • To enable efficient deployment of RNNs on resource-constrained edge devices.

Main Methods:

  • A structured pruning method based on neuron selection is proposed, removing independent neurons from RNNs.
  • Two sets of binary random variables (gates/switches) are introduced for input and hidden neurons.
  • The optimization problem is solved by minimizing the L0 norm of the weight matrix.

Main Results:

  • The proposed method demonstrates effectiveness in language modeling and machine reading comprehension tasks.
  • Significant practical speedup (nearly 20x) was achieved during inference for a language model on the Penn TreeBank dataset.
  • The method maintained performance levels comparable to state-of-the-art pruning competitors.

Conclusions:

  • The structured pruning method offers a practical solution for reducing RNN size and computation.
  • This approach facilitates the deployment of efficient RNNs on edge devices without compromising performance.
  • The neuron selection strategy effectively achieves practical speedups, showcasing promising results for real-world applications.