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

An architecture for fully integrated large scale neural networks

M D Binns1, F J Clough, S C Garth

  • 1Department of Engineering, Cambridge University, UK.

International Journal of Neural Systems
|December 1, 1993
PubMed
Summary

This study introduces a novel amorphous silicon technology for creating large-scale neural networks. This approach overcomes crystalline silicon limitations, enabling simpler fabrication of advanced artificial intelligence hardware.

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

Deep Model Families for EEG-Based Multi-Class Dementia Classification.

International journal of neural systems·2026
Same journal

Toward Efficient and Generalizable Text Dataset Distillation via a Dual-Agent Large Language Model Framework.

International journal of neural systems·2026
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

Area of Science:

  • Materials Science
  • Electrical Engineering
  • Computer Science

Background:

  • Large neural networks are crucial for engineering applications but face fabrication challenges with crystalline silicon.
  • Existing methods struggle with the scalability and integration required for complex neural network hardware.

Purpose of the Study:

  • To propose and describe a novel architecture for neural network hardware using amorphous silicon technology.
  • To overcome the limitations of crystalline silicon in fabricating large-scale neural networks.

Main Methods:

  • Utilizing amorphous silicon photoresistors for storing synaptic weights, with up to 100 million per plate.
  • Employing an external light source for individual photoresistor adjustment, configuring them as programmable resistors.

Related Experiment Videos

  • Leveraging polysilicon and amorphous silicon processing compatibility for integrated photosensors, analogue neural networks, and neurons on a single glass substrate.

    • Main Results:

    • Demonstrated a simple, elegant, and easily fabricated neural network architecture.
    • Successfully integrated photosensors, analogue neural networks, and neurons on a single glass substrate.
    • Alleviated input interface problems through integrated photosensors.

    Conclusions:

    • Amorphous silicon technology offers a viable solution for fabricating large-scale neural network hardware.
    • The proposed architecture simplifies design and fabrication, paving the way for more advanced AI systems.
    • This approach addresses key limitations in current neural network hardware development.