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 generalized divergence measure for nonnegative matrix factorization.

Raul Kompass1

  • 1FU Berlin, Institut für Mathematik und Informatik, 14152 Berlin, Germany. kompass@inf.fu-berlin.de

Neural Computation
|February 15, 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 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

Researchers introduced a new parametric divergence measure, generalizing existing metrics like quadratic error and Kullback-Leibler divergence. This new method enhances nonnegative matrix factorization convergence speed and offers locally optimal solutions.

Area of Science:

  • Mathematics
  • Computer Science
  • Machine Learning

Background:

  • Nonnegative matrix factorization (NMF) is a widely used dimensionality reduction technique.
  • Existing divergence measures, such as quadratic error and Kullback-Leibler divergence, have limitations in certain NMF applications.
  • The multiplicative update rules by Lee and Seung (2001) are standard for solving NMF problems.

Purpose of the Study:

  • To introduce a general parametric divergence measure for NMF.
  • To demonstrate that this new measure can improve convergence speed and solution quality in NMF.
  • To provide a theoretical framework for the new measure and its associated update rules.

Main Methods:

  • Developed a general parametric divergence measure.
  • Parametrically generalized the multiplicative update rules of Lee and Seung (2001).

Related Experiment Videos

  • Conducted numeric simulations to evaluate performance.
  • Provided a proof of convergence using an auxiliary function from the expectation-maximization framework.

    • Main Results:

    • The proposed parametric divergence measure includes quadratic error and Kullback-Leibler divergence as special cases.
    • The generalized update rules lead to locally optimal solutions for NMF with the new cost function.
    • Numeric simulations indicate potential improvements in convergence speed for quadratic distance.
    • A convergence proof was established, drawing parallels with the expectation-maximization framework.

    Conclusions:

    • The new parametric divergence measure offers a flexible alternative for NMF.
    • The associated update rules provide an effective method for achieving locally optimal NMF solutions.
    • The findings contribute to the theoretical understanding and practical application of NMF.