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

Robust L1 principal component analysis and its Bayesian variational inference.

Junbin Gao1

  • 1School of Computer Science, Charles Sturt University, Bathurst, NSW, Australia. jbgao@csu.edu.au

Neural Computation
|November 30, 2007
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

Human-Like Multimodal Fake News Detection via Reflective Summarization and Large-Small Model Collaboration.

IEEE transactions on neural networks and learning systems·2026
Same author

Hybrid graph attention learning with pseudo-label guided adaptive evolution.

Neural networks : the official journal of the International Neural Network Society·2026
Same author

Attribute-Topology Cross-Frequency Aligned Graph Neural Networks for Homophilic and Heterophilic Graphs in Node Classification.

IEEE transactions on neural networks and learning systems·2026
Same author

Piezoelectric nanomotors for active cartilage regeneration of osteoarthritis via ultrasonic vibration and water splitting.

Biomaterials·2025
Same author

Multimodal Knowledge Graph Completion by Cross-Modal Interaction With Similarity Enhancing and Difference Embracing.

IEEE transactions on neural networks and learning systems·2025
Same author

Mitigating inconsistencies in GWAS follow-up analyses with LocusCompare2.

Nature genetics·2025
Same journal

A Model-Free Reinforcement Learning Implementation of Decision Making Under Uncertainty by Sequential Sampling.

Neural computation·2026
Same journal

DROP: Distributional and Regular Optimism and Pessimism for Reinforcement Learning.

Neural computation·2026
Same journal

Hierarchical Active Inference Using Successor Representations.

Neural computation·2026
Same journal

W-Kernel and Its Principal Space for Frequentist Evaluation of Bayesian Estimators.

Neural computation·2026
Same journal

A Hidden Markov Model-Inspired Sequence Classification Method for Hyperdimensional Computing.

Neural computation·2026
Same journal

Sparse Graphical Modeling for Electrophysiological Phase-Based Connectivity Using Circular Statistics.

Neural computation·2026
See all related articles

This study introduces a robust probabilistic L1-PCA model, replacing Gaussian noise with the Laplacian (L1) distribution for improved outlier resistance. A variational approximation enables effective parameter inference for this robust principal component analysis method.

Area of Science:

  • Machine Learning
  • Statistical Modeling
  • Data Analysis

Background:

  • Traditional Principal Component Analysis (PCA) assumes Gaussian noise, making it sensitive to outliers.
  • Robustness in statistical modeling is crucial for reliable analysis of real-world datasets.
  • Outliers can significantly distort results in standard PCA, necessitating alternative approaches.

Purpose of the Study:

  • To develop a robust probabilistic L1-PCA model that is resilient to data outliers.
  • To implement an effective parameter inference method for the proposed L1-PCA model.
  • To leverage the heavy-tail properties of the Laplacian distribution for enhanced data analysis.

Main Methods:

  • Replaced the conventional Gaussian noise assumption with the Laplacian (L1) distribution in a probabilistic PCA framework.

Related Experiment Videos

  • Utilized a variational approximation scheme for efficient inference of model parameters.
  • Expressed the L1-PCA model as a marginalized model over infinite Gaussian density superpositions.

    • Main Results:

    • The proposed L1-PCA model demonstrates increased robustness against data outliers due to the L1 distribution's heavy tails.
    • The variational approximation enables tractable Bayesian inference for the L1-PCA model.
    • The model effectively handles datasets with non-Gaussian noise characteristics.

    Conclusions:

    • The L1-PCA model offers a more robust alternative to standard PCA for datasets containing outliers.
    • Variational inference provides an effective computational approach for parameter estimation in L1-PCA.
    • This work contributes to robust statistical modeling and machine learning techniques.