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

Automating parallel implementation of neural learning algorithms.

O F Rana1

  • 1Department of Computer Science, University of Wales, Cardiff, UK.

International Journal of Neural Systems
|September 30, 2000
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 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

This study presents a design scheme for implementing neural learning algorithms on parallel hardware. It uses a specialized Neural Language and Petri net models to optimize concurrency and performance for applications like Kohonen and backpropagation.

Area of Science:

  • Artificial Intelligence
  • Computational Neuroscience
  • Parallel Computing

Background:

  • Neural learning algorithms utilize interconnected processing units with specific update functions and learning rules.
  • Exploiting concurrency in neural algorithms requires effective implementation strategies for parallel architectures.
  • Existing methods often limit global interaction or synchronous updates, hindering parallel processing.

Purpose of the Study:

  • To describe a design scheme for translating neural learning algorithms into parallel implementations.
  • To enable efficient execution on parallel machines (e.g., using PVM or MPI) and programmable logic (FPGAs).
  • To facilitate the verification and performance modeling of neural learning rules through Petri net models.

Main Methods:

Related Experiment Videos

  • A specialized Neural Language is used to describe the neural learning algorithm.
  • Automatic construction of Petri net (PN) models from the Neural Language description for verification and performance analysis.
  • A scheduling and mapping strategy translates the PN model onto a multiprocessor template.
  • Demonstration using Kohonen and backpropagation learning rules on workstation clusters and dedicated parallel machines with PVM.

    • Main Results:

    • The proposed design scheme effectively translates neural learning algorithms for parallel execution.
    • Petri net models allow for the analysis of synchronization, resource sharing, and concurrency within learning rules.
    • Successful implementation of Kohonen and backpropagation algorithms on diverse parallel hardware demonstrated the technique's versatility.

    Conclusions:

    • The described design scheme provides a systematic approach to implementing neural learning algorithms on parallel systems.
    • The use of Neural Language and Petri nets enhances the analysis and optimization of concurrency in neural learning.
    • This methodology offers a viable solution for efficient deployment of complex neural learning algorithms on modern parallel architectures.