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

Selection of generative models in classification.

Guillaume Bouchard1, Gilles Celeux

  • 1Xerox Research Centre Europe, 6, Ch de maupertuis, 38240 Meylan, France. Guillaume.Bouchard@xrce.xerox.com

IEEE Transactions on Pattern Analysis and Machine Intelligence
|March 29, 2006
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

Roles for RpoS in survival of Escherichia coli during protozoan predation and in reduced moisture conditions highlight its importance in soil environments.

FEMS microbiology letters·2017
Same author

A model selection criterion for model-based clustering of annotated gene expression data.

Statistical applications in genetics and molecular biology·2015
Same author

Latent IBP Compound Dirichlet Allocation.

IEEE transactions on pattern analysis and machine intelligence·2015
Same author

Co-expression analysis of high-throughput transcriptome sequencing data with Poisson mixture models.

Bioinformatics (Oxford, England)·2015
Same author

Comparing Model Selection and Regularization Approaches to Variable Selection in Model-Based Clustering.

Journal de la Societe francaise de statistique (2009)·2014
Same author

Fast and efficient estimation of individual ancestry coefficients.

Genetics·2014
Same journal

Relation DETR+: Exploring Explicit Position Relation Prior for Dense Prediction.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

RBF++: Quantifying and Optimizing Reasoning Boundaries across Measurable and Unmeasurable Capabilities for Chain-of-Thought Reasoning.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

CAFE: Cross-View Adaptive Fusion and Cluster Center Enhancement for Robust Multi-View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Ethics-Aware Safe Reinforcement Learning for Rare-Event Risk Control in Interactive Urban Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Learning Shape Anchors for Holistic Indoor Scene Understanding.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

A new Bayesian Entropy Criterion (BEC) aids generative model selection for classification tasks. BEC effectively minimizes classification error, offering a computationally efficient alternative to cross-validation.

Area of Science:

  • Machine Learning
  • Statistical Modeling
  • Pattern Recognition

Background:

  • Traditional model selection criteria focus on model fit, not classification accuracy.
  • Selecting generative models for supervised classification requires evaluating predictive performance.

Purpose of the Study:

  • To introduce the Bayesian Entropy Criterion (BEC) for generative model selection in supervised classification.
  • To provide a computationally efficient alternative to cross-validation for model selection.
  • To evaluate the performance of BEC against existing criteria.

Main Methods:

  • Development of the Bayesian Entropy Criterion (BEC) based on minimizing integrated classification entropy.
  • Analysis of the asymptotic behavior of the BEC criterion.

Related Experiment Videos

  • Empirical evaluation using simulated and real-world datasets.

    • Main Results:

    • BEC directly optimizes for low classification error rates.
    • BEC demonstrates superior performance compared to the Bayesian Information Criterion (BIC) for classification error minimization.
    • BEC achieves performance comparable to computationally expensive cross-validation.

    Conclusions:

    • The Bayesian Entropy Criterion (BEC) is a viable and efficient method for selecting generative models in supervised classification.
    • BEC offers a practical approach for optimizing predictive accuracy in classification tasks.
    • BEC presents a valuable alternative to traditional model selection techniques when classification performance is paramount.