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

Cascaded redundancy reduction

V R de Sa1, G E Hinton

  • 1Department of Computer Science, University of Toronto, Ontario, Canada.

Network (Bristol, England)
|December 23, 1998
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

Spatio-temporal analysis of error-related brain activity in active and passive brain-computer interfaces.

Brain computer interfaces (Abingdon, England)·2020
Same author

Glove-TalkII--a neural-network interface which maps gestures to parallel formant speech synthesizer controls.

IEEE transactions on neural networks·2008
Same author

Reducing the dimensionality of data with neural networks.

Science (New York, N.Y.)·2006
Same author

Computation by neural networks.

Nature neuroscience·2000
Same author

SMEM algorithm for mixture models.

Neural computation·2000
Same author

Variational learning for switching state-space models.

Neural computation·2000
Same journal

Enhancing IoT security: A Creative Swagger Optimization algorithm for DDoS defence.

Network (Bristol, England)·2026
Same journal

Parametric optimization for electrical discharge diamond grinding (EDDG) system using dual approach.

Network (Bristol, England)·2025
Same journal

A novel lung cancer diagnosis model using hybrid convolution (2D/3D)-based adaptive DenseUnet with attention mechanism.

Network (Bristol, England)·2025
Same journal

Hybrid optimization enabled Eff-FDMNet for Parkinson's disease detection and classification in federated learning.

Network (Bristol, England)·2025
Same journal

AI-driven plant disease detection with tailored convolutional neural network.

Network (Bristol, England)·2025
Same journal

Layer modified residual Unet++ for speech enhancement using Aquila Black widow optimizer algorithm.

Network (Bristol, England)·2025
See all related articles

We present a novel hierarchical generative model for binary data. This method incrementally adds hidden units to minimize information, simplifying model training.

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Computational Neuroscience

Background:

  • Hierarchical generative models are crucial for understanding complex data structures.
  • Stochastic binary logistic units offer a flexible framework for modeling binary data ensembles.
  • Incremental model construction can improve computational efficiency and scalability.

Purpose of the Study:

  • To introduce a method for incrementally building a hierarchical generative model for binary data vectors.
  • To detail the composition of the model using stochastic, binary, logistic units.
  • To explain the optimization objective of minimizing information required to describe data vectors.

Main Methods:

  • The model is constructed by adding hidden units incrementally.

Related Experiment Videos

  • The objective is to minimize the information needed to describe data vectors.
  • Both top-down generative weights and bottom-up recognition weights are utilized.
  • A recognition model determines hidden unit states, simplifying weight searching despite underestimation.

    • Main Results:

    • The proposed method allows for the incremental construction of a hierarchical generative model.
    • The model effectively utilizes stochastic, binary, logistic units for data representation.
    • The integration of recognition weights simplifies the training process for new hidden units.

    Conclusions:

    • The described method provides an efficient approach to building hierarchical generative models for binary data.
    • The incremental addition of hidden units and the use of recognition weights offer a practical solution for complex modeling tasks.
    • This work contributes to advancements in generative modeling and machine learning techniques.