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

Higher-order probabilistic perceptrons as Bayesian inference engines.

J W Clark1, K A Gernoth, S Dittmar

  • 1McDonnell Center for the Space Sciences and Department of Physics, Washington University, St. Louis, Missouri 63130, USA.

Physical Review. E, Statistical Physics, Plasmas, Fluids, and Related Interdisciplinary Topics
|April 24, 2002
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

A crown-group cnidarian from the Ediacaran of Charnwood Forest, UK.

Nature ecology & evolution·2022
Same author

Crizotinib in ROS1-rearranged advanced non-small-cell lung cancer (NSCLC): updated results, including overall survival, from PROFILE 1001.

Annals of oncology : official journal of the European Society for Medical Oncology·2019
Same author

Phase II study of gemcitabine, oxaliplatin in combination with panitumumab in KRAS wild-type unresectable or metastatic biliary tract and gallbladder cancer.

British journal of cancer·2014
Same author

Merging of Landau levels in a strongly interacting two-dimensional electron system in silicon.

Physical review letters·2014
Same author

The activity of phosphonic and phosphinic acids on Treponema pallidum.

Antibiotics & chemotherapy (Northfield, Ill.)·2014
Same author

Clinical benefit of continuing ALK inhibition with crizotinib beyond initial disease progression in patients with advanced ALK-positive NSCLC.

Annals of oncology : official journal of the European Society for Medical Oncology·2014
Same journal

Efficient Monte Carlo simulations using a shuffled nested Weyl sequence random number generator.

Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics·2002
Same journal

Spatiotemporal dynamics of electromagnetic pulses in saturating nonlinear optical media with normal group velocity dispersion.

Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics·2002
Same journal

Soliton-breather reaction pathways.

Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics·2002
Same journal

Calculation of electromagnetic properties of regular and random arrays of metallic and dielectric cylinders.

Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics·2002
Same journal

Electromagnetic convective cells in a nonuniform dusty plasma.

Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics·2002
Same journal

Stability of neural networks and solitons of field theory.

Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics·2002
See all related articles

This study links Bayes optimal classifiers to higher-order probabilistic perceptrons. These networks can learn classification statistics and predict nuclide stability, matching conventional methods.

Area of Science:

  • Machine Learning
  • Computational Physics
  • Statistical Inference

Background:

  • Bayes optimal classifiers are powerful for binary classification tasks.
  • Conventional multilayer perceptrons (MLPs) use pairwise connections.
  • Higher-order connections in neural networks are less explored.

Purpose of the Study:

  • To establish a structural link between Bayes optimal classifiers and higher-order probabilistic perceptrons.
  • To develop training algorithms for these perceptrons.
  • To evaluate their performance in a nuclear physics application.

Main Methods:

  • Formulating a two-layer perceptron with higher-order couplings (up to K).
  • Developing variational training algorithms for approximating posterior probabilities.

Related Experiment Videos

  • Applying the model to discriminate between stable and unstable nuclides.

    • Main Results:

    • Demonstrated an explicit connection between Bayes optimal classifiers and higher-order perceptrons.
    • Showcased the ability of these perceptrons to learn classification statistics.
    • Achieved comparable predictive performance to conventional MLPs in nuclide stability prediction.

    Conclusions:

    • Higher-order probabilistic perceptrons offer a principled approach to classification.
    • These networks can effectively model complex data distributions.
    • They provide a viable alternative to conventional MLPs, particularly in domains like nuclear physics.