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

Pattern Categorization and Generalization with a Virtual Neuromolecular Architecture.

Michael Conrad1, Jong Chen Chen

  • 1Wayne State University, USA

Neural Networks : the Official Journal of the International Neural Network Society
|January 1, 1997
PubMed
Summary

This study introduces a novel neuromolecular computing architecture for evolutionary learning. This system utilizes neuron-like modules and evolutionary algorithms to achieve complex pattern recognition and neurocontrol.

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

Bulletin of mathematical biology·2016
Same author

Diagnostic implications of an elevated troponin in the emergency department.

Disease markers·2015
Same author

Diagnostic performance of copeptin in patients with acute nontraumatic chest pain: BWH-TIMI ED chest pain study.

Clinical cardiology·2014
See all related articles

Area of Science:

  • Computational Neuroscience
  • Artificial Intelligence
  • Biomolecular Computing

Background:

  • Real neurons integrate signals via molecular processes, including second messengers and cytoskeleton-membrane interactions.
  • Existing computational models often lack the complexity of biological neural signal integration.

Purpose of the Study:

  • To develop a multilevel neuromolecular computing architecture for evolutionary learning.
  • To create specialized dynamic pattern processors using evolutionary search.
  • To enable pattern recognition and neurocontrol through memory manipulation.

Main Methods:

  • Utilizing a network of neuron-like modules with cellular automata for internal dynamics.
  • Implementing evolutionary search algorithms to generate a repertoire of pattern processors.

Related Experiment Videos

  • Employing memory manipulation algorithms to select and combine processors for specific tasks.
  • Incorporating two layers of cytoskeletally controlled neurons and two layers of reference neurons.

    • Main Results:

    • The developed network successfully performs complex pattern categorization tasks.
    • The system demonstrates a balance between specificity and generalization in pattern recognition.
    • Evolutionary learning occurs at intraneuronal levels (cytoskeletal structures, protein locations, connectivity).

    Conclusions:

    • The neuromolecular computing architecture offers a robust platform for evolutionary learning.
    • The system effectively achieves pattern recognition and neurocontrol through a combination of evolutionary and memory-based strategies.
    • This approach models neuronal signal integration using molecular process hypotheses.