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 principal component analysis by self-organizing rules based on statistical physics approach.

L Xu1, A L Yuille

  • 1Dept. of Comput. Sci., Chinese Univ. of Hong Kong, Shatin.

IEEE Transactions on Neural Networks
|January 1, 1995
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

Erratum to "Annotated normal CT data of the abdomen for deep learning: Challenges and strategies for implementation" [Diagn. Interv. Imaging. 101 (2020) 35-44].

Diagnostic and interventional imaging·2020
Same author

Differentiating autoimmune pancreatitis from pancreatic ductal adenocarcinoma with CT radiomics features.

Diagnostic and interventional imaging·2020
Same author

Deep lessons learned: Radiology, oncology, pathology, and computer science experts unite around artificial intelligence to strive for earlier pancreatic cancer diagnosis.

Diagnostic and interventional imaging·2019
Same author

Annotated normal CT data of the abdomen for deep learning: Challenges and strategies for implementation.

Diagnostic and interventional imaging·2019
Same author

Deformable templates for face recognition.

Journal of cognitive neuroscience·2013
Same author

Scaling theorems for zero crossings.

IEEE transactions on pattern analysis and machine intelligence·2011
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

This study introduces robust principal component analysis (PCA) rules using statistical physics to effectively handle data outliers. The new methods significantly improve PCA performance in the presence of noisy data.

Area of Science:

  • Statistical Physics
  • Machine Learning
  • Data Analysis

Background:

  • Principal Component Analysis (PCA) is a common dimensionality reduction technique.
  • Standard PCA methods and existing self-organizing rules perform poorly with data outliers.
  • Robustness to outliers is crucial for reliable PCA applications.

Purpose of the Study:

  • To develop robust principal component analysis (PCA) methods using statistical physics.
  • To explicitly address and mitigate the impact of outliers in PCA.
  • To enhance the performance of PCA for various data analysis tasks.

Main Methods:

  • Applied statistical physics principles to formulate generalized energy functions for PCA.
  • Incorporated binary decision fields with prior distributions to handle outliers.

Related Experiment Videos

  • Derived self-organizing rules for robust PCA from marginal distributions of Gibbs distributions.

    • Main Results:

    • The proposed robust PCA rules demonstrate significant resistance to outliers.
    • Outperformed standard PCA and existing self-organizing PCA methods in the presence of outliers.
    • Successfully performed PCA-like tasks including identifying principal component vectors and subspaces.

    Conclusions:

    • The developed robust PCA rules offer a significant improvement over existing methods when outliers are present.
    • These novel rules provide a more reliable approach for dimensionality reduction in noisy datasets.
    • The statistical physics framework offers a powerful approach for developing robust machine learning algorithms.