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 hierarchical method for finding optimal architecture and weights using evolutionary least square based learning.

Ranadhir Ghosh1, Brijesh Verma

  • 1School of Information Technology, Griffith University, PMB 50, Gold Coast Mail Center, QLD 9726,Australia. r.ghosh@gu.edu.au

International Journal of Neural Systems
|March 15, 2003
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

Investigational New Drug Enabling Nonclinical Study of Xenogeneic Life-Supporting Porcine Kidneys With 10 Gene Edits (10 GE) in a Nonhuman Primate Test System.

Xenotransplantation·2026
Same author

Monocyte Distribution Width (MDW) in Patients with COVID-19: An Indicator of Disease Severity.

Indian journal of hematology & blood transfusion : an official journal of Indian Society of Hematology and Blood Transfusion·2023
Same author

Multicluster Class-Balanced Ensemble.

IEEE transactions on neural networks and learning systems·2020
Same author

In Field Fruit Sizing Using A Smart Phone Application.

Sensors (Basel, Switzerland)·2018
Same author

On-Tree Mango Fruit Size Estimation Using RGB-D Images.

Sensors (Basel, Switzerland)·2017
Same author

Novel layered clustering-based approach for generating ensemble of classifiers.

IEEE transactions on neural networks·2011
Same journal

Latent Space Projections and Atlases, a Cautionary Tale in Deep Neuroimaging using Autoencoders.

International journal of neural systems·2026
Same journal

Transformer-Based Anomaly Detection for Neurodegenerative Screening in MRI Images.

International journal of neural systems·2026
Same journal

Discrete Wavelet Convolution for Learnable Time-Frequency Representation with Application to Seizure Prediction.

International journal of neural systems·2026
Same journal

Automatic Seizure Detection using Hierarchical Spectral-Temporal Feature Learning with an Imbalance-Aware Transformer.

International journal of neural systems·2026
Same journal

Pyramid Vision Transformer-Enhanced Conformer Network for Epileptic Seizure Recognition Using MultiChannel EEG Signals.

International journal of neural systems·2026
Same journal

A Time-Frequency Decoupled Contrastive Learning Framework for Electroencephalography-Based Parkinson's Disease Diagnosis.

International journal of neural systems·2026
See all related articles

This study introduces a novel evolutionary least squares algorithm (GALS) to optimize neural network architecture and weights. GALS efficiently finds optimal solutions for complex problems, improving classification accuracy.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Neuroscience

Background:

  • Optimizing neural network architecture and weights is crucial for performance.
  • Existing methods face challenges in complex, multidimensional search spaces.
  • Developing efficient algorithms for neural network design remains an active research area.

Purpose of the Study:

  • To present a novel combined methodology for determining optimal neural network architecture and weights.
  • To introduce the Growing Architecture-based Least Squares (GALS) algorithm.
  • To evaluate GALS performance against existing methods on benchmark datasets.

Main Methods:

  • Implementing a combined methodology using an evolutionary least squares-based algorithm (GALS).

Related Experiment Videos

  • Integrating weight updating heuristics with a growing architecture model to determine hidden neuron count.
  • Addressing challenges such as probabilistic solution space initialization and fitness breaking.

    • Main Results:

    • GALS successfully identified appropriate neural network architectures and weights.
    • The approach was applied to the XOR problem, 10-bit odd parity, and real-world datasets (handwriting, breast cancer, heart disease).
    • Comparative analysis demonstrated competitive classification accuracy and time complexity.

    Conclusions:

    • The proposed GALS approach offers an effective method for optimizing neural networks.
    • Combining evolutionary least squares with growing architectures efficiently determines network parameters.
    • GALS shows promise for solving complex problems in machine learning and related fields.