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 variational method for learning sparse and overcomplete representations.

M Girolami1

  • 1Laboratory of Computing and Information Science, Helsinki University of Technology, Finland.

Neural Computation
|October 25, 2001
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

Two-Photon Sub-Bandgap Photocurrent in Surface-Nanotextured Black Diamond Films for Solar Energy Conversion.

ACS photonics·2025
Same author

Management of symptomatic spine metastases: A multidisciplinary approach based flow-chart.

Revista espanola de cirugia ortopedica y traumatologia·2023
Same author

Management of symptomatic spine metastases: A multidisciplinary approach based flow-chart.

Revista espanola de cirugia ortopedica y traumatologia·2023
Same author

A high-resolution neutron spectroscopic camera for the SPARC tokamak based on the Jet European Torus deuterium-tritium experience.

The Review of scientific instruments·2022
Same author

Carbon Ion Radiotherapy for sacral chordoma: a systematic review of the literature.

Journal of biological regulators and homeostatic agents·2020
Same author

Novel concepts and nanostructured materials for thermionic-based solar and thermal energy converters.

Nanotechnology·2020
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 an expectation-maximization algorithm for learning sparse and overcomplete data representations. It enables analytic marginalization for improved data likelihood and basis coefficient inference.

Area of Science:

  • Machine Learning
  • Signal Processing
  • Computational Neuroscience

Background:

  • Learning sparse and overcomplete representations is crucial for understanding complex data.
  • Existing methods often struggle with heavy-tailed distributions and computational efficiency.

Purpose of the Study:

  • To develop a novel expectation-maximization (EM) algorithm for learning sparse and overcomplete data representations.
  • To leverage variational approximations for handling heavy-tailed distributions.
  • To enable analytic marginalization for improved model inference.

Main Methods:

  • The study proposes an EM algorithm utilizing variational approximation to heavy-tailed distributions, specifically the Laplacian.
  • A rigorous lower bound on the sparse prior distribution is derived.

Related Experiment Videos

  • This enables analytic marginalization of a lower bound on the data likelihood.

    • Main Results:

    • The developed algorithm effectively learns overcomplete basis vectors.
    • It facilitates the inference of the most probable basis coefficients.
    • The approach provides a robust method for sparse representation learning.

    Conclusions:

    • The proposed EM algorithm offers an effective solution for learning sparse and overcomplete representations from data.
    • The use of variational approximation and derived lower bounds enhances the analytic tractability and performance of the learning process.
    • This work contributes to advancements in unsupervised feature learning and data representation.