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

Bayesian Gaussian process classification with the EM-EP algorithm.

Hyun-Chul Kim1, Zoubin Ghahramani

  • 1Department of Industrial and Management Engineering, Pohang University of Science and Technology, Nam-gu, Pohang, Republic of China. grass@postech.ac.kr

IEEE Transactions on Pattern Analysis and Machine Intelligence
|November 17, 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

The effect of a structured injury prevention program on injury severity and type in elite Adolescent and Young Adult Taekwondo athletes.

Physical therapy in sport : official journal of the Association of Chartered Physiotherapists in Sports Medicine·2026
Same author

Training-related injury patterns and return-to-sports in elite fencing athletes: a prospective cohort study.

The Physician and sportsmedicine·2026
Same author

Whole-Brain Task fMRI Decoding Using Stage-Wise Residual-Optimized 3D ConvNeXt With Layer-Global Response Normalization.

IEEE journal of biomedical and health informatics·2025
Same author

Transcranial focused ultrasound to nucleus accumbens reverses fentanyl-induced conditioned place preference in rats.

PloS one·2025
Same author

Genetic diversity of Olive flounder (Paralichthys olivaceus) and the impact of selective breeding on Korean populations.

PloS one·2025
Same author

PCR sequencing and re-evaluation of pathologic findings revealed invasive fungal sinusitis caused by Malassezia restricta.

Diagnostic microbiology and infectious disease·2025
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

This study introduces an efficient EM-EP algorithm for Gaussian Process Classifiers (GPCs). The new method effectively learns latent functions and kernel hyperparameters, matching or exceeding existing GPC and SVM performance.

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Statistical Modeling

Background:

  • Gaussian Process Classifiers (GPCs) are Bayesian kernel methods for classification.
  • Inferring latent functions and hyperparameters is crucial for GPC performance.
  • Expectation Propagation (EP) offers a method for inferring the latent function posterior.

Purpose of the Study:

  • To develop an efficient approximate EM algorithm for GPCs.
  • To enable robust learning of latent functions and kernel hyperparameters.
  • To extend the EM-EP algorithm for multiclass classification problems.

Main Methods:

  • An approximate EM algorithm (EM-EP) is proposed, building upon the EP approach for GPCs.
  • The EM-EP algorithm jointly learns the latent function and hyperparameters.

Related Experiment Videos

  • A multiclass extension of the EM-EP algorithm is derived.

    • Main Results:

    • The EM-EP algorithm demonstrates practical convergence.
    • It provides an efficient Bayesian framework for hyperparameter learning in GPCs.
    • Experimental results show EM-EP performance is comparable or superior to existing GPC and SVM methods.

    Conclusions:

    • The EM-EP algorithm offers an effective and efficient Bayesian approach for GPCs.
    • The multiclass extension enhances the applicability of GPCs.
    • This method provides a strong alternative for classification tasks.