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

Nonnegative matrix factorization with Gaussian process priors.

Mikkel N Schmidt1, Hans Laurberg

  • 1Department of Informatics and Mathematical Modelling, Technical University of Denmark, Richard Petersens Plads, DTU-Building 321, 2800 Lyngby, Denmark. mns@imm.dtu.dk

Computational Intelligence and Neuroscience
|May 10, 2008
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

Probabilistic PARAFAC2.

Entropy (Basel, Switzerland)·2024
Same author

Graph neural network interatomic potential ensembles with calibrated aleatoric and epistemic uncertainty on energy and forces.

Physical chemistry chemical physics : PCCP·2023
Same author

Nitroaromatic explosives' detection and quantification using an attention-based transformer on surface-enhanced Raman spectroscopy maps.

The Analyst·2023
Same author

Raman spectrum matching with contrastive representation learning.

The Analyst·2022
Same author

Uncovering Cortical Units of Processing From Multi-Layered Connectomes.

Frontiers in neuroscience·2022
Same author

Using connectomics for predictive assessment of brain parcellations.

NeuroImage·2021
Same journal

RETRACTION: Multidimensional Heterogeneous Network Link Adaptation Based on Mobile Environment.

Computational intelligence and neuroscience·2026
Same journal

RETRACTION: Framework to Segment and Evaluate Multiple Sclerosis Lesion in MRI Slices Using VGG-UNet.

Computational intelligence and neuroscience·2026
Same journal

RETRACTION: Facial Emotion Recognition Using a Novel Fusion of Convolutional Neural Network and Local Binary Pattern in Crime Investigation.

Computational intelligence and neuroscience·2026
Same journal

RETRACTION: Automatic Intelligent System Using Medical of Things for Multiple Sclerosis Detection.

Computational intelligence and neuroscience·2026
Same journal

RETRACTION: Intangible Cultural Heritage Reproduction and Revitalization: Value Feedback, Practice, and Exploration Based on the IPA Model.

Computational intelligence and neuroscience·2026
Same journal

RETRACTION: CNN Based Multiclass Brain Tumor Detection Using Medical Imaging.

Computational intelligence and neuroscience·2025
See all related articles

We developed a new method to incorporate prior knowledge into nonnegative matrix factorization (NMF) using Gaussian process priors. This approach enables NMF decompositions that align with expected factor distributions like sparseness and smoothness.

Area of Science:

  • Machine Learning
  • Statistical Modeling
  • Signal Processing

Background:

  • Nonnegative matrix factorization (NMF) is a widely used dimensionality reduction technique.
  • Incorporating prior knowledge into NMF can improve decomposition accuracy and interpretability.
  • Existing methods for prior knowledge integration in NMF are limited.

Purpose of the Study:

  • To present a general method for integrating prior knowledge into NMF using Gaussian process priors.
  • To enable NMF decompositions that reflect prior assumptions about factor distributions.
  • To demonstrate the method's utility in a practical application.

Main Methods:

  • Formulating NMF with Gaussian process priors, linking nonnegative factors to underlying Gaussian processes.

Related Experiment Videos

  • Specifying the Gaussian process using its covariance function to encode prior knowledge.
  • Developing an algorithm to find NMF decompositions consistent with prior knowledge.

    • Main Results:

    • The proposed method successfully incorporates prior knowledge such as sparseness, smoothness, and symmetries into NMF.
    • Demonstrated the method's effectiveness using an example from chemical shift brain imaging.
    • Achieved NMF decompositions that better reflect expected data characteristics.

    Conclusions:

    • Gaussian process priors offer a flexible and powerful framework for incorporating prior knowledge into NMF.
    • The presented method enhances the interpretability and accuracy of NMF decompositions.
    • This approach has potential applications in various fields requiring structured data analysis.