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

Kolmogorov's Theorem Is Relevant.

Věra Kůrková1

  • 1Institute of Computer Science, Czechoslovak Academy of Sciences, P. O. Box 5, 182 07 Prague 8, Czechoslovakia.

Neural Computation
|June 7, 2019
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

The rise of <i>Candidozyma auris</i> in Czechia: three clades, prosthetic joint infection and fluconazole resistance development, 2022 to 2024.

Euro surveillance : bulletin Europeen sur les maladies transmissibles = European communicable disease bulletin·2025
Same author

Approximation of classifiers by deep perceptron networks.

Neural networks : the official journal of the International Neural Network Society·2023
Same author

Exploratory drilling: how to set up, carry out, and evaluate a seroprevalence study.

Casopis lekaru ceskych·2020
Same author

Translation-Invariant Kernels for Multivariable Approximation.

IEEE transactions on neural networks and learning systems·2020
Same author

Some insights from high-dimensional spheres: Comment on "The unreasonable effectiveness of small neural ensembles in high-dimensional brain" by Alexander N. Gorban et al.

Physics of life reviews·2019
Same author

Classification by Sparse Neural Networks.

IEEE transactions on neural networks and learning systems·2019
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

Kolmogorov

Area of Science:

  • Mathematics
  • Computer Science
  • Artificial Intelligence

Background:

  • Kolmogorov's theorem provides a theoretical basis for representing complex functions using simpler ones.
  • Neural networks are powerful function approximators used in artificial intelligence.
  • Understanding the theoretical underpinnings of neural networks is crucial for their development.

Purpose of the Study:

  • To demonstrate the relevance of Kolmogorov's theorem to neural network theory.
  • To adapt Kolmogorov's theorem for approximating functions using neural network components.
  • To establish theoretical bounds on the complexity of neural networks.

Main Methods:

  • Applying Kolmogorov's theorem to the architecture of neural networks.
  • Approximating single-variable functions using compositions of affine transformations and sigmoidal functions.

Related Experiment Videos

  • Deriving upper bounds for the number of hidden units required.
  • Main Results:

    • Kolmogorov's theorem is applicable to understanding neural network representations.
    • A version of the theorem allows approximating functions with specific neural network building blocks.
    • An upper estimate for the number of hidden units in neural networks was derived.

    Conclusions:

    • The study bridges theoretical mathematics and practical neural network design.
    • The findings offer insights into the representational capacity of neural networks.
    • The derived estimates can guide the efficient design of neural network architectures.