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

A unified framework for connectionist systems.

R M Golden1

  • 1Department of Psychology, Stanford University, CA 94305.

Biological Cybernetics
|January 1, 1988
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

Use of a neural network to predict stone growth after shock wave lithotripsy.

Urology·1998
Same author

Predictors of general quality of life in patients with benign prostate hyperplasia or prostate cancer.

The Journal of urology·1997
Same author

Cruisin' the information superhighway. Learn how to enter the world of cyberEMS on the Internet, online services and BBS systems.

Emergency medical services·1995
Same author

Evaluation of medicolegal investigators' suspicions and positive toxicology findings in 100 drug deaths.

Journal of forensic sciences·1986
Same author

Vocoder filter design: practical considerations.

The Journal of the Acoustical Society of America·1968
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 frames neural network pattern classification within statistics, interpreting responses as probability distributions. This statistical view enables improved learning algorithms and network evaluation for various connectionist models.

Area of Science:

  • Artificial Intelligence
  • Computational Neuroscience
  • Statistical Learning Theory

Background:

  • Connectionist models, such as neural networks, are widely used for pattern classification.
  • Existing approaches often lack a rigorous statistical foundation for understanding network behavior.
  • A unified statistical framework is needed to analyze and improve these models.

Purpose of the Study:

  • To develop a statistical framework for understanding pattern classification in connectionist models.
  • To interpret a network's internal state as a subjective probability distribution.
  • To enable the design of learning algorithms and statistical tests based on this framework.

Main Methods:

  • Deriving a connectionist network's subjective beliefs about its environment.

Related Experiment Videos

  • Representing these beliefs as a "subjective" probability distribution.
  • Interpreting stimulus classification as finding the most probable response based on this distribution.
  • Utilizing maximum likelihood estimation for algorithm design and statistical testing.

    • Main Results:

    • A novel statistical framework is established for connectionist pattern classification.
    • The framework provides a probabilistic interpretation of network responses.
    • Learning algorithms can be designed and analyzed using maximum likelihood estimation.
    • Statistical tests can be developed for evaluating and comparing network architectures.

    Conclusions:

    • The proposed statistical framework offers a principled approach to understanding and developing connectionist models.
    • This framework is broadly applicable to various existing neural network architectures.
    • It facilitates the rigorous analysis and optimization of pattern classification in artificial intelligence.