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

Two coding strategies for bidirectional associative memory.

Y F Wang1, J R Cruz, J R Mulligan

  • 1Dept. of Electr. and Comput. Eng., California Univ., Irvine, CA.

IEEE Transactions on Neural Networks
|January 1, 1990
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

The First Report of Cucumber Mosaic Virus Infecting Water Chestnut in China.

Plant disease·2019
Same author

First Report of Alternaria alternata Causing Black Leaf Spot of Rubber Tree in China.

Plant disease·2019
Same author

[Adenocarcinoma arising from traditional serrated adenoma in the jejunum: report of a case].

Zhonghua bing li xue za zhi = Chinese journal of pathology·2019
Same author

Measurement of the Electron Antineutrino Oscillation with 1958 Days of Operation at Daya Bay.

Physical review letters·2019
Same author

Measurement of the Absolute Branching Fraction of the Inclusive Semileptonic Λ_{c}^{+} Decay.

Physical review letters·2019
Same author

Search for Subsolar-Mass Ultracompact Binaries in Advanced LIGO's First Observing Run.

Physical review letters·2018
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

This study enhances discrete bidirectional associative memory (BAM) with multiple training and dummy augmentation for improved data recall. These methods ensure reliable retrieval of single or multiple trained pairs, advancing BAM

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Neural Networks

Background:

  • The original discrete bidirectional associative memory (BAM) model by B. Kosko (1987) has limitations in recalling multiple trained pairs.
  • Enhancing BAM encoding strategies is crucial for improving associative memory performance in artificial intelligence applications.

Purpose of the Study:

  • To present novel enhancements to the discrete bidirectional associative memory (BAM) encoding strategy.
  • To introduce and validate two new concepts: multiple training and dummy augmentation.
  • To improve the reliability and completeness of data recall in BAM systems.

Main Methods:

  • Multiple training: A method designed to guarantee recall of a single trained pair under specific initial data conditions.
  • Dummy augmentation: A technique that ensures recall of all trained pairs by attaching dummy data.

Related Experiment Videos

  • Mathematical analysis and computer simulations to validate the proposed methods and compare them with the original Kosko strategy.

    • Main Results:

    • Multiple training demonstrated improved recall of multiple pairs compared to the original Kosko strategy in simulations.
    • Dummy augmentation guarantees recall of all trained pairs when applicable.
    • A sufficient condition for correlation matrices to ensure training pair energies are local minima was discussed.

    Conclusions:

    • The proposed multiple training and dummy augmentation strategies significantly enhance the performance of discrete bidirectional associative memory (BAM).
    • These enhancements lead to more reliable and complete recall of stored information.
    • The study provides theoretical underpinnings and practical validation for improved BAM encoding strategies.