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

Learning and coding of concepts in neural networks.

Y Salu

    Bio Systems
    |January 1, 1985
    PubMed
    Summary
    This summary is machine-generated.

    The brain analyzes complex environments by recognizing familiar concepts using a small library of basic elements. This study introduces neural network models for these "classifying boxes" and evaluates their concept-learning mechanisms via simulations.

    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

    An improved method for localizing electric brain dipoles.

    IEEE transactions on bio-medical engineering·1990
    Same author

    Models of neural novelty detectors, with similarities to cerebral cortex.

    Bio Systems·1988
    Same author

    Theoretical models and computer simulations of neural learning systems.

    Journal of theoretical biology·1984
    Same author

    A computerized system for localizing sources of cardiac activation.

    Computers and biomedical research, an international journal·1984
    Same author

    Computer simulations of learning in neural systems.

    Computers and biomedical research, an international journal·1983
    Same author

    Orthogonal surface lead recordings of His-Purkinje activity: comparison of actual and simulated waveforms.

    IEEE transactions on bio-medical engineering·1983

    Area of Science:

    • Cognitive neuroscience
    • Computational neuroscience
    • Artificial intelligence

    Background:

    • The brain must interpret vast environmental stimuli using a limited set of basic concepts.
    • Many essential concepts are learned rather than innate, adapting to arbitrary surroundings.
    • Understanding concept formation is key to understanding brain function.

    Purpose of the Study:

    • To introduce models of neural networks, termed "classifying boxes," designed to identify familiar concepts within stimuli.
    • To propose and evaluate mechanisms by which these networks acquire their concept libraries.
    • To compare the effectiveness of different concept-learning strategies through computational simulations.

    Main Methods:

    • Development of neural network models (classifying boxes) for stimulus analysis.

    Related Experiment Videos

  • Proposal of potential mechanisms for the establishment of concept libraries within these networks.
  • Computer simulations to compare and evaluate the performance of suggested mechanisms.

    • Main Results:

    • Demonstration of functional "classifying box" models capable of identifying familiar concepts.
    • Comparative analysis of different concept-learning mechanisms through simulation data.
    • Evaluation of the efficiency and efficacy of various library-building strategies.

    Conclusions:

    • Neural network models can effectively identify familiar concepts from stimuli.
    • Simulations provide insights into how concept libraries are formed and learned.
    • The proposed models and mechanisms offer a framework for understanding brain-based concept acquisition.