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 Experiment Videos

A new algorithm to design compact two-hidden-layer artificial neural networks.

M M Islam1, K Murase

  • 1Department of Human and Artificial Intelligence Systems, Fukui University, Japan.

Neural Networks : the Official Journal of the International Neural Network Society
|November 23, 2001
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Potassium Imbalance in Patients with Hypertension and Acute Haemorrhagic Stroke in a Tertiary Level Hospital of Bangladesh.

Mymensingh medical journal : MMJ·2026
Same author

Effect of Nd: YAG Laser Capsulotomy Size on Visual Acuity and Posterior Segment Outcomes: A 6-Months Follow-Up Study.

Mymensingh medical journal : MMJ·2026
Same author

Effect of Intraoperative Infiltration of Local Anaesthetic Agent on Post-tonsillectomy Pain.

Mymensingh medical journal : MMJ·2025
Same author

Impact of Adenotonsillectomy in Children Suffering From Recurrent Upper Respiratory Tract Infections.

Mymensingh medical journal : MMJ·2025
Same author

Adverse Events Following Sinopharm Covid-19 Vaccination among Medical Students.

Mymensingh medical journal : MMJ·2025
Same author

Knowledge, Attitude and Practice about Covid-19 among the Adult Population of a Sub-urban Community in Bangladesh.

Mymensingh medical journal : MMJ·2025
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
Same journal

IGFD-Net: Illumination-guided frequency decoupling for polarization image fusion.

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

Multiple-Strategies dung beetle optimizer and its applications in engineering optimization and bankruptcy prediction.

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

Aggregating global-scale pixel-wise forgery cues within a graph.

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

Finite-Time intermittent control for secure synchronization of Neutral-Type stochastic delayed neural networks under aperiodic DoS attacks.

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

The Cascade Neural Network Design Algorithm (CNNDA) efficiently creates compact artificial neural networks (ANNs). This new method optimizes both network performance and training speed for complex problems.

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Machine Learning

Background:

  • Artificial Neural Networks (ANNs) are powerful tools for complex pattern recognition.
  • Designing compact ANNs with optimal generalization and training time remains a challenge.
  • Existing algorithms often require extensive computational resources and time.

Purpose of the Study:

  • To introduce a novel algorithm, the Cascade Neural Network Design Algorithm (CNNDA), for designing compact, two-hidden-layer ANNs.
  • To optimize both the generalization ability and training time of ANNs through an automated design process.
  • To reduce the computational cost and accelerate the training of ANNs.

Main Methods:

  • The CNNDA employs a combination of constructive and pruning algorithms.

Related Experiment Videos

  • Bounded fan-ins for hidden nodes are utilized to enhance generalization.
  • A novel training approach involves temporarily freezing input weights of hidden nodes to reduce computational load.
  • The algorithm automatically determines ANN architecture and connection weights.

    • Main Results:

    • The CNNDA successfully produced compact ANNs across various benchmark datasets.
    • Experimental results demonstrated superior generalization ability compared to other algorithms.
    • The algorithm achieved significantly shorter training times.
    • Tested on problems including cancer, diabetes, and character recognition.

    Conclusions:

    • The CNNDA is an effective algorithm for designing efficient and high-performing ANNs.
    • It offers a significant improvement in balancing network compactness, generalization, and training speed.
    • The CNNDA presents a promising approach for practical applications requiring rapid and accurate ANN deployment.