The universal approximation property. Characterization, construction, representation, and existence

DOI10.1007/s10472-020-09723-1OpenAlexW3108320661MaRDI QIDQ2043428

Anastasis Kratsios

Publication date: 2 August 2021

Published in: Annals of Mathematics and Artificial Intelligence (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1910.03344

zbMATH Keywords

topological transitivityuniform approximationcomposition operatorsuniversal approximationdeep learningconstrained approximation

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Multidimensional problems (41A63) Notions of recurrence and recurrent behavior in topological dynamical systems (37B20) Applications of functional analysis in numerical analysis (46N40) Geometric embeddings of metric spaces (30L05)

Related Items (7)

A singular Riemannian geometry approach to deep neural networks. I: Theoretical foundations ⋮ Neural networks in Fréchet spaces ⋮ Time discretization in the solution of parabolic PDEs with ANNs ⋮ Universal regular conditional distributions via probabilistic transformers ⋮ Rethinking the ill-posedness of the spectral function reconstruction -- why is it fundamentally hard and how artificial neural networks can help ⋮ Discovering first principle of behavioural change in disease transmission dynamics by deep learning ⋮ Separability Versus Robustness of Orlicz Spaces: Financial and Economic Perspectives

Uses Software

• Keras
• hydra+
• Wasserstein GAN

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Hydra: A method for strain-minimizing hyperbolic embedding of network- and distance-based data
• Dynamical properties of weighted composition operators on the space of smooth functions
• Linear chaos
• On embedding uniform and topological spaces
• Subreflexive normed linear space
• Linear extension operators between spaces of Lipschitz maps and optimal transport
• Topological transitivity and mixing of composition operators
• Multilayer feedforward networks are universal approximators
• Topological properties of the set of functions generated by neural networks of fixed size
• Geodesic regression and the theory of least squares on Riemannian manifolds
• Weighted spaces of vector-valued continuous functions
• La dualité dans les espaces \((\mathcal F)\) et \((\mathcal{LF})\)
• A survey on Lipschitz-free Banach Spaces
• Locally Convex Spaces
• Some generalizations of the Stone Duality Theorem
• A Refinement of a Mixed Sparse Effective Nullstellensatz
• M-structure and the Banach-Stone theorem
• Universal approximation bounds for superpositions of a sigmoidal function
• Hamiltonian Systems and Transformation in Hilbert Space
• Lipschitz Algebras
• Lipschitz-free Banach spaces
• Deep hedging
• Dynamics of weighted composition operators on function spaces defined by local properties
• Deep optimal stopping
• Structure of categories
• A logical calculus of the ideas immanent in nervous activity
• Approximation by superpositions of a sigmoidal function
• Riemannian geometry and geometric analysis

This page was built for publication: The universal approximation property. Characterization, construction, representation, and existence