Deep neural network classifier for multidimensional functional data

DOI10.1111/SJOS.12660MaRDI QIDQ6140335zbMATH OpenOpenAlexFDO

Authors Shuoyang Wang, Guanqun Cao, Zuofeng Shang

Publication date 2 January 2024

Published in Scandinavian Journal of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2205.08592

zbMATH Keywords

functional data analysisfunctional classificationfunctional neural networksminimax excess misclassification riskmultidimensional functional data

Mathematics Subject Classification ID

Statistics (62-XX)

Abstract: We propose a new approach, called as functional deep neural network (FDNN), for classifying multi-dimensional functional data. Specifically, a deep neural network is trained based on the principle components of the training data which shall be used to predict the class label of a future data function. Unlike the popular functional discriminant analysis approaches which rely on Gaussian assumption, the proposed FDNN approach applies to general non-Gaussian multi-dimensional functional data. Moreover, when the log density ratio possesses a locally connected functional modular structure, we show that FDNN achieves minimax optimality. The superiority of our approach is demonstrated through both simulated and real-world datasets.

Cites work

• scientific article; zbMATH DE number 47282 (Why is no real title available?)
• Achieving near perfect classification for functional data
• An extension of Fisher's discriminant analysis for stochastic processes
• Classification using censored functional data
• Classification with minimax fast rates for classes of Bayes rules with sparse representation
• Componentwise classification and clustering of functional data
• Error bounds for approximations with deep ReLU networks
• Fast convergence rates of deep neural networks for classification
• Fast learning rates for plug-in classifiers
• Globally robust inference for the location and simple linear regression models
• High Dimensional Linear Discriminant Analysis: Optimality, Adaptive Algorithm and Missing Data
• Introduction to Functional Data Analysis
• Introduction to nonparametric estimation
• Nonparametric regression using deep neural networks with ReLU activation function
• OUP accepted manuscript
• On deep learning as a remedy for the curse of dimensionality in nonparametric regression
• On the Use of Reproducing Kernel Hilbert Spaces in Functional Classification
• Optimal aggregation of classifiers in statistical learning.
• Optimal approximation of piecewise smooth functions using deep ReLU neural networks
• Optimal nonparametric inference via deep neural network
• Smooth discrimination analysis
• Tensor product space ANOVA models.
• Theoretical foundations of functional data analysis, with an introduction to linear operators
• Transfer learning for nonparametric classification: minimax rate and adaptive classifier

Cited in

(7)

• Nonlinear principal component analysis for two-dimensional functional data using neural networks
• CovNet: Covariance Networks for Functional Data on Multidimensional Domains
• Functional data analysis using deep neural networks
• Detection and localization of changes in a panel of densities
• Deep Neural Network for Functional Graphical Models Structure Learning
• Supervised Manifold Learning for Functional Data
• Functional data analysis applications in medicine: a systematic review

This page was built for publication: Deep neural network classifier for multidimensional functional data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6140335)