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

Biologically relevant neural network architectures for support vector machines.

Magnus Jändel1

  • 1Swedish Defence Research Agency, SE 164 90 Stockholm, Sweden.

Neural Networks : the Official Journal of the International Neural Network Society
|October 16, 2013
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Neural Circuits01:25

Neural Circuits

3.0K
Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
3.0K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Natural evolution of neural support vector machines.

Advances in experimental medicine and biology·2011
Same author

A neural support vector machine.

Neural networks : the official journal of the International Neural Network Society·2010
Same author

Signatures of depression in non-stationary biometric time series.

Computational intelligence and neuroscience·2009
See all related articles

This study explores neural network models for one-shot learning, finding that competitive queuing memory (CQM) architectures effectively implement support vector machines (SVMs). Four feasible neural models were identified, with a ν-SVM showing a simple implementation in bisymmetric networks.

Area of Science:

  • Computational Neuroscience
  • Machine Learning
  • Cognitive Science

Background:

  • Biological organisms exhibit remarkable one-shot learning capabilities.
  • Support Vector Machines (SVMs) are powerful classification algorithms.
  • Neural network architectures offer a framework for modeling biological learning.

Purpose of the Study:

  • To investigate neural network architectures implementing SVMs for modeling perceptual one-shot learning.
  • To adapt SVM training rules for neural computation.
  • To identify feasible neural models for biological learning.

Main Methods:

  • Consideration of various SVM algorithms (max-margin, 1-norm, 2-norm, ν-SVM).
  • Derivation of SVM training rules suitable for neural computation.
Keywords:
Competitive queuing memoryNeural networkPerceptual learningSupport vector machine

Related Experiment Videos

  • Examination of Competitive Queuing Memory (CQM)-based neural architectures.
  • Evaluation of sixty-four distinct CQM-based architectures.

    • Main Results:

    • Competitive Queuing Memory (CQM) is identified as optimal for storing and retrieving support vectors.
    • Four feasible neural network architectures were found suitable for biological modeling.
    • A ν-SVM learning rule demonstrated a simple and natural implementation within bisymmetric architectures.

    Conclusions:

    • CQM-like structures may encode skilled action sequences.
    • Bisymmetry in motor systems suggests a link to perceptual pattern recognition.
    • Trainable pattern recognition in low-level perception may have evolved from internalized motor programs.