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 hybrid learning network for shift-invariant recognition.

R Wang1

  • 1Engineering Department, Harvey Mudd College, Claremont, CA 91711, USA. ruye_wang@hmc.edu

Neural Networks : the Official Journal of the International Neural Network Society
|October 30, 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

[Osteoarthropathy in pregnancy].

Zhonghua fu chan ke za zhi·2001
Same author

Glomerular laminin isoform transitions: errors in metanephric culture are corrected by grafting.

American journal of physiology. Renal physiology·2001
Same author

[The relationship of vascular endothelial growth factor and angiogenesis to the progression of gastric carcinoma].

Zhonghua bing li xue za zhi = Chinese journal of pathology·2001
Same author

Monte Carlo dose calculations of beta-emitting sources for intravascular brachytherapy: a comparison between EGS4, EGSnrc, and MCNP.

Medical physics·2001
Same author

Tmtacn, tacn, and triammine complexes of (eta 6-arene)OsII: syntheses, characterizations, and photosubstitution reactions (tmtacn = 1,4,7-trimethyl-1,4,7-triazacyclononane; tacn = 1,4,7-triazacyclononane).

Inorganic chemistry·2001
Same author

A panel immunoblot using co-incubated monoclonal antibodies for identification of melanoma cells.

Journal of immunological methods·2001
Same journal

Dynamic analysis and reliable mechanical optimization application of ring HNN effected with a memristive neuron.

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

DAFF-Net: A detection and search method for small-scale low surface brightness galaxies.

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

Quasi-synchronization for complex networks with hybrid pinning intermittent control.

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

Physics-encoded convolutional neural operators for parametric PDEs: A convergence-guaranteed framework via pre-computed kernel fields.

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

This study introduces a novel neural network for invariant visual pattern recognition, achieving object identification regardless of location, orientation, or scale. The biologically plausible model uses a hybrid unsupervised learning algorithm for robust and generic pattern recognition applications.

Area of Science:

  • Computational Neuroscience
  • Artificial Intelligence
  • Computer Vision

Background:

  • Biological visual systems exhibit remarkable invariant object recognition capabilities.
  • Current computational models often struggle to replicate this invariance across varying geometric attributes.

Purpose of the Study:

  • To present a generic neural network architecture and learning algorithm for invariant visual pattern recognition.
  • To develop a biologically plausible computational model for understanding visual processing.
  • To create a robust algorithm applicable to diverse recognition tasks.

Main Methods:

  • A multi-layer hierarchical neural network with spatially arranged input layer groups.
  • Nodes in higher layers receive lateral and vertical inputs.

Related Experiment Videos

  • An unsupervised hybrid learning algorithm combining competitive and Hebbian learning.

    • Main Results:

    • The network architecture and hybrid learning facilitate emergent invariant recognition at the output layer.
    • Demonstrated robustness and generic applicability of the proposed approach.
    • The model provides a simplified yet plausible computational account of biological invariant recognition.

    Conclusions:

    • The proposed neural network and hybrid learning algorithm effectively achieve invariant visual pattern recognition.
    • The model serves as a biologically plausible computational tool for studying visual systems.
    • The generic and robust nature of the algorithm supports its application in various practical recognition problems.