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 associative memory that can form hypotheses: a phase-coded neural network

N Kunstmann1, C Hillermeier, B Rabus

  • 1Institut für Medizinische Optik, Theoretische Biophysik, Ludwig-Maximilians-Universität München, Germany.

Biological Cybernetics
|January 1, 1994
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 "Hot-Solvent/Cold-Solute" Problem Revisited.

Journal of chemical theory and computation·2015
Same author

Type A-trichothecenes - Quantitative analysis using LC-MS and occurrence in Austrian maize and oats.

Mycotoxin research·2013
Same author

Investigation on the biodegradability of mycotoxins nivalenol (NIV) and deoxynivalenol (DON) in a rusitec fermentor and their monitoring by HPLC/MS.

Mycotoxin research·2013
Same author

Study on biodegradation of some A- and B-trichothecenes and ochratoxin A by use of probiotic microorganisms.

Mycotoxin research·2013
Same author

[Not Available].

Mycotoxin research·2013
Same author

Polarization effects stabilize bacteriorhodopsin's chromophore binding pocket: a molecular dynamics study.

The journal of physical chemistry. B·2009
Same journal

Harmonic memory in phasor neural networks.

Biological cybernetics·2026
Same journal

Correction: Decreased spinal inhibition leads to undiversified locomotor patterns.

Biological cybernetics·2026
Same journal

Foundational issues of network models in biology.

Biological cybernetics·2026
Same journal

Dynamical mechanisms for coordinating long-term working memory based on the precision of spike-timing in cortical neurons.

Biological cybernetics·2026
Same journal

Distinct dopaminergic spike-timing-dependent plasticity rules are suited to different functional roles.

Biological cybernetics·2026
Same journal

Fluctuation-response relations for a two-stage population of spiking neurons stimulated by common noise.

Biological cybernetics·2026
See all related articles

This study introduces a novel neural network model extending Hopfield nets with neuron phases. This innovation enables associative memories to generate multiple classification hypotheses, overcoming limitations of traditional models.

Area of Science:

  • Computational neuroscience
  • Artificial intelligence
  • Machine learning

Background:

  • Nonlinear associative memories, like Hopfield nets, use attractor dynamics, converging to a single stored pattern.
  • Conventional neural networks struggle with classification tasks requiring multiple potential outputs.
  • Existing models neglect detailed neural activity correlations, limiting their representational capacity.

Purpose of the Study:

  • To develop an associative memory capable of generating multiple classification hypotheses.
  • To extend conventional neural network architectures by incorporating oscillatory dynamics and neural phases.
  • To address the limitations of single-answer convergence in attractor-based memory systems.

Main Methods:

  • Inspired by correlation theory, introduced additional dynamical variables representing neural phases.

Related Experiment Videos

  • Assumed an oscillatory time structure for neural firing, akin to neural clocks.
  • Implemented the extended architecture into a self-organizing network based on a feature map.

    • Main Results:

    • The novel associative memory successfully forms hypotheses of classification.
    • Explicitly modeling neural phases captures detailed correlations previously neglected.
    • The system demonstrates the ability to propose multiple plausible classifications for a given input.

    Conclusions:

    • The proposed model overcomes the single-hypothesis limitation of traditional associative memories.
    • Incorporating neural phases and oscillatory dynamics enhances representational power.
    • This approach offers a new paradigm for classification tasks in artificial neural networks.