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 analog feedback associative memory.

A Atiya1, Y S Abu-Mostafa

  • 1Dept. of Electr. Eng., California Inst. of Technol., Pasadena, CA.

IEEE Transactions on Neural Networks
|January 1, 1993
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

Prevalence and characteristics of cigarillo smoking in Canada: results from the Canadian Tobacco Use Monitoring Survey.

Public health·2018
Same author

Recognitive aspects of moment invariants.

IEEE transactions on pattern analysis and machine intelligence·2012
Same author

Image normalization by complex moments.

IEEE transactions on pattern analysis and machine intelligence·2011
Same author

Financial model calibration using consistency hints.

IEEE transactions on neural networks·2008
Same author

An algorithmic approach to adaptive state filtering using recurrent neural networks.

IEEE transactions on neural networks·2008
Same author

Density estimation and random variate generation using multilayer networks.

IEEE transactions on neural networks·2008
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

A new method enhances analog vector storage in continuous-time Hopfield networks using hidden units. This approach guarantees storage of multiple vectors, overcoming previous limitations.

Area of Science:

  • Computational Neuroscience
  • Machine Learning
  • Artificial Neural Networks

Background:

  • The Hopfield network is a recurrent neural network model used for associative memory.
  • Storing analog vectors in continuous-time Hopfield networks presents stability and capacity challenges.
  • Existing models have limitations in the number and type of vectors that can be reliably stored.

Purpose of the Study:

  • To develop a method for reliably storing analog vectors in continuous-time Hopfield networks.
  • To address limitations of the standard Hopfield model regarding vector storage capacity.
  • To ensure stored analog vectors correspond to asymptotically stable equilibria.

Main Methods:

  • A novel architecture incorporating visible and hidden units in a circular fashion was proposed.

Related Experiment Videos

  • Mathematical proofs were developed to demonstrate storage guarantees for a two-layer network.
  • A learning algorithm was designed to set equilibrium locations and ensure asymptotic stability.

    • Main Results:

    • The proposed architecture, particularly the two-layer case, guarantees storage of analog vectors.
    • Storage capacity is linked to the number of hidden neurons, exceeding previous limits.
    • The developed learning algorithm effectively adjusts equilibria and ensures their stability.

    Conclusions:

    • The new method significantly improves analog vector storage in continuous-time Hopfield networks.
    • The inclusion of hidden units and a specific architecture overcomes prior storage limitations.
    • The approach is validated by simulations, confirming its effectiveness and stability guarantees.