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 correlation significance learning scheme for auto-associative memories

D L Lee1, W J Wang

  • 1Institute of Electrical Engineering, National Central University, Taiwan, R.O.C.

International Journal of Neural Systems
|December 1, 1995
PubMed
Summary

This study introduces correlation significance to enhance memory capacity in asynchronous auto-associative networks. By unequally weighting neural connections, the method maximizes attraction regions, improving memory recall and storage.

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

[A case report of small cell lung cancer with acute pancreatitisand and intestinal pseudo-obstruction as initial manifestations].

Zhonghua nei ke za zhi·2026
Same author

[Study on the effect of high-intensity interval exercise on ulcerative colitis in mice].

Zhonghua yu fang yi xue za zhi [Chinese journal of preventive medicine]·2026
Same author

[Variant frequency of <i>GJB2</i> c.109G>A (p.Val37Ile) in Chinese patients with hearing loss: a systematic review and Meta-analysis].

Zhonghua er bi yan hou tou jing wai ke za zhi = Chinese journal of otorhinolaryngology head and neck surgery·2026
Same author

[Primary prevention study of preimplantation genetic testing for <i>GJB2</i> gene mutation families].

Zhonghua er bi yan hou tou jing wai ke za zhi = Chinese journal of otorhinolaryngology head and neck surgery·2026
Same author

[Development of a deep learning model for predicting adverse cardiovascular events in patients with acute coronary syndrome based on retinal fundus images].

Zhonghua xin xue guan bing za zhi·2026
Same author

[Analysis of five-year outcomes and risk factors for insufficient weight loss after laparoscopic sleeve gastrectomy in obese patients].

Zhonghua wei chang wai ke za zhi = Chinese journal of gastrointestinal surgery·2026

Area of Science:

  • Artificial Intelligence
  • Computational Neuroscience
  • Machine Learning

Background:

  • Traditional outer product rules in neural networks lead to suboptimal attraction regions.
  • Maximizing attraction regions is crucial for improving the storage capacity of associative memories.

Purpose of the Study:

  • Introduce a novel concept, correlation significance, to enhance asynchronous auto-associative memory.
  • Develop a method to maximize attraction regions around stored vectors (attractors).

Main Methods:

  • Devised a rule for unequally weighting correlations between different components of stored patterns.
  • Designed the neural network's connection matrix (T) using a gradient descent approach.
  • Constructed an exponential error function to monitor learning progress and storage success.

Related Experiment Videos

Main Results:

  • The proposed method effectively maximizes attraction regions in asynchronous auto-associative memory.
  • Computer simulations demonstrate the enhanced efficiency and capability of the new scheme.
  • The error function allows direct examination of successfully stored vectors during learning.

Conclusions:

  • Correlation significance offers a novel approach to optimize neural network memory.
  • The gradient descent method with unequal weighting significantly improves attractor stability and memory capacity.
  • This scheme presents a promising advancement for asynchronous auto-associative memory systems.