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 Bayesian approach to joint feature selection and classifier design.

Balaji Krishnapuram1, Alexander J Hartemink, Lawrence Carin

  • 1Department of Electrical Engineering, Duke University, Durham, NC 27708-0291, USA. balaji@ee.duke.edu

IEEE Transactions on Pattern Analysis and Machine Intelligence
|March 4, 2005
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

Gradient Importance Learning for Incomplete Observations.

... International Conference on Learning Representations·2026
Same author

Comprehensive profiling of chromatin occupancy dynamics through the cell cycle.

Nucleic acids research·2026
Same author

Genome-wide nucleosome and transcription factor responses to genetic perturbations reveal chromatin-mediated mechanisms of transcriptional regulation.

Genome research·2025
Same author

Capturing Actionable Dynamics with Structured Latent Ordinary Differential Equations.

Proceedings of machine learning research·2025
Same author

Enabling Counterfactual Survival Analysis with Balanced Representations.

ACM CHIL 2021 : proceedings of the 2021 ACM Conference on Health, Inference, and Learning : April 8-9, 2021, Virtual Event. ACM Conference on Health, Inference, and Learning (2021 : Online)·2025
Same author

Inferring differential protein binding from time-series chromatin accessibility data.

Bioinformatics advances·2025
Same journal

Benchmarking the Robustness of Autonomous Driving to Environmental Illusions: A Lane Perception Perspective.

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

Learning Topology-Aware Representations via Test-Time Adaptation for Anomaly Segmentation.

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

TraGraph-GS: Trajectory Graph-based Gaussian Splatting for Arbitrary Large-Scale Scene Rendering.

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

SWIFT: A Small-World Interaction Framework for Flow-Aware Trajectory Prediction in Autonomous Driving.

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

HardFlow: Hard-Constrained Sampling for Flow-Matching Models Via Trajectory Optimization.

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

Industrial Brain: Self-Evolving Neuro-Symbolic Autonomy with Causal Resilience for Cyber-Physical Systems.

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

This study introduces a Bayesian method for selecting optimal nonlinear classifiers and relevant features simultaneously. The approach ensures parsimonious feature selection and high classification accuracy using heavy-tailed priors and an expectation-maximization algorithm.

Area of Science:

  • Machine Learning
  • Statistical Inference
  • Computational Statistics

Background:

  • Traditional classification methods often struggle with high-dimensional data and selecting relevant features.
  • Sparse Bayesian learning offers a principled framework for model selection and regularization.
  • Support Vector Machines (SVMs) are effective but can be computationally intensive and less interpretable.

Purpose of the Study:

  • To develop a Bayesian approach for simultaneous learning of nonlinear classifiers and feature subsets.
  • To promote sparsity in basis function and feature utilization for improved model interpretability and efficiency.
  • To provide an efficient computational method for estimating model parameters.

Main Methods:

  • A Bayesian framework employing heavy-tailed priors to induce sparsity in basis functions and features.

Related Experiment Videos

  • Development of an expectation-maximization (EM) algorithm for computing maximum a posteriori (MAP) estimates.
  • Extension of existing sparse Bayesian classification techniques, linking them to Bayesian SVM counterparts.

    • Main Results:

    • The proposed method effectively performs parsimonious feature selection, identifying the most relevant predictor variables.
    • Demonstrated excellent classification accuracy across various synthetic and benchmark datasets.
    • The EM algorithm provides an efficient means to compute MAP estimates for model parameters.

    Conclusions:

    • The Bayesian approach offers a robust and efficient method for nonlinear classification with integrated feature selection.
    • Heavy-tailed priors effectively regularize the model, leading to sparse solutions and improved generalization.
    • This work advances sparse Bayesian learning and provides a powerful alternative to existing classification techniques.