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

Functional multi-layer perceptron: a non-linear tool for functional data analysis.

Fabrice Rossi1, Brieuc Conan-Guez

  • 1CEREMADE, UMR CNRS 7534, Université Paris-IX Dauphine, Place du Maréchal de Lattre de Tassigny, 75016 Paris, France. fabrice.rossi@inria.fr

Neural Networks : the Official Journal of the International Neural Network Society
|January 15, 2005
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

Predicting local malaria exposure using a Lasso-based two-level cross validation algorithm.

PloS one·2017
Same author

A bag-of-paths framework for network data analysis.

Neural networks : the official journal of the International Neural Network Society·2017
Same author

Fast algorithm and implementation of dissimilarity self-organizing maps.

Neural networks : the official journal of the International Neural Network Society·2006
Same journal

Anchor-based disentanglement framework for incremental multi-view clustering.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Complex-valued amplitude-phase interference modeling for adversarially robust classification.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

TraNce: Type-aware hypergraph neural network with biological mediators for drug repositioning.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Decentralized ADMM for factorization-based Low-rank matrix estimation.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Memristive neuromorphic circuit design inspired by the neural mechanisms of conditioned fear.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Q-learning based asynchronous Boolean control networks stabilization with data loss.

Neural networks : the official journal of the International Neural Network Society·2026
See all related articles

This study introduces functional multi-layer perceptrons (MLP) for complex data. These functional MLPs demonstrate comparable expressive power and statistically sound parameter estimation to traditional MLPs.

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Functional Data Analysis

Background:

  • Multi-layer perceptrons (MLPs) are foundational neural networks.
  • Extending MLPs to handle functional data is an active research area.
  • Classical MLP theory provides a benchmark for network capabilities.

Purpose of the Study:

  • To introduce and analyze a natural extension of MLPs for functional inputs.
  • To establish theoretical guarantees for functional MLPs.
  • To evaluate the practical performance of functional MLPs.

Main Methods:

  • Theoretical extension of classical MLP principles to functional spaces.
  • Derivation of universal approximation theorems for functional MLPs.
  • Development of consistency results for parameter estimation.

Related Experiment Videos

  • Empirical validation using simulated and real-world functional datasets.

    • Main Results:

    • Functional MLPs possess universal approximation capabilities.
    • The expressive power of functional MLPs is equivalent to numerical MLPs.
    • Parameter estimation for functional MLPs is statistically consistent.
    • The proposed functional MLP model shows strong performance on diverse data.

    Conclusions:

    • Functional MLPs offer a powerful and theoretically grounded approach for analyzing functional data.
    • The findings extend the applicability of deep learning to a broader range of data types.
    • The model's satisfactory performance validates its utility in practical applications.