A deep network construction that adapts to intrinsic dimensionality beyond the domain

DOI10.1016/j.neunet.2021.06.004arXiv2008.02545OpenAlexW3171226399MaRDI QIDQ6054952

Alexander Cloninger, Timo Klock

Publication date: 28 September 2023

Published in: Neural Networks (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2008.02545

zbMATH Keywords

approximation theorycurse of dimensionalitycomposite functionsdeep neural networksnoisy manifold models

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Multidimensional problems (41A63)

Related Items (7)

Shared subspace-based radial basis function neural network for identifying ncRNAs subcellular localization ⋮ Drift estimation for a multi-dimensional diffusion process using deep neural networks ⋮ Deep nonparametric estimation of intrinsic data structures by chart autoencoders: generalization error and robustness ⋮ Side effects of learning from low-dimensional data embedded in a Euclidean space ⋮ Neural network approximation and estimation of classifiers with classification boundary in a Barron class ⋮ Stable recovery of entangled weights: towards robust identification of deep neural networks from minimal samples ⋮ Deep neural networks can stably solve high-dimensional, noisy, non-linear inverse problems

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Hölder-Lipschitz norms and their duals on spaces with semigroups, with applications to earth mover's distance
• On the tractability of multivariate integration and approximation by neural networks
• Learning and approximation by Gaussians on Riemannian manifolds
• Random projections of smooth manifolds
• Approximation and estimation bounds for artificial neural networks
• Multilayer feedforward networks are universal approximators
• Provable approximation properties for deep neural networks
• Optimal nonlinear approximation
• Optimal global rates of convergence for nonparametric regression
• Approximation properties of a multilayered feedforward artificial neural network
• Limitations of the approximation capabilities of neural networks with one hidden layer
• Theory of deep convolutional neural networks. II: Spherical analysis
• A direct approach for function approximation on data defined manifolds
• Optimal approximation of piecewise smooth functions using deep ReLU neural networks
• Nonlinear approximation via compositions
• Theory of deep convolutional neural networks: downsampling
• Function approximation by deep networks
• Nonparametric regression using deep neural networks with ReLU activation function
• Error bounds for approximations with deep ReLU networks
• Universality of deep convolutional neural networks
• On deep learning as a remedy for the curse of dimensionality in nonparametric regression
• The reach, metric distortion, geodesic convexity and the variation of tangent spaces
• Manifold reconstruction using tangential Delaunay complexes
• Finding the homology of submanifolds with high confidence from random samples
• Deep vs. shallow networks: An approximation theory perspective
• Curvature Measures
• Universal approximation bounds for superpositions of a sigmoidal function
• Neural Networks for Localized Approximation
• Bounds on rates of variable-basis and neural-network approximation
• Comparison of worst case errors in linear and neural network approximation
• High-Dimensional Probability
• A Review on Dimension Reduction
• Optimal Approximation with Sparsely Connected Deep Neural Networks
• Deep neural networks for rotation-invariance approximation and learning
• Minimax Manifold Estimation
• Approximation by superpositions of a sigmoidal function

This page was built for publication: A deep network construction that adapts to intrinsic dimensionality beyond the domain