Universal approximation property of fully convolutional neural networks with zero padding

DOI10.1016/J.ACHA.2025.101833MaRDI QIDQ6858026zbMATH OpenFDO

Authors Geonho Hwang, Myungjoo Kang

Publication date 13 February 2026

Published in Applied and Computational Harmonic Analysis (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/J.ACHA.2025.101833

zbMATH Keywords

convolutional neural networkdeep narrow networkfully convolutional neural networkminimum depthuniversal approximation theoremzero padding

Mathematics Subject Classification ID

Approximation by other special function classes (41A30) Artificial neural networks and deep learning (68T07)

Cites work

• Approximation by superpositions of a sigmoidal function
• Approximation properties of deep ReLU CNNs
• Equivalence of approximation by convolutional neural networks and fully-connected networks
• Multilayer feedforward networks are universal approximators
• On the convergence of infinite compositions of entire functions
• The Stone-Weierstrass Theorem
• Theory of deep convolutional neural networks: downsampling
• Universal approximations of invariant maps by neural networks
• Universality of deep convolutional neural networks

This page was built for publication: Universal approximation property of fully convolutional neural networks with zero padding

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6858026)