English
Related papers

Related papers: Approximation Rates for Neural Networks with Gener…

200 papers

We propose two approaches of locally adaptive activation functions namely, layer-wise and neuron-wise locally adaptive activation functions, which improve the performance of deep and physics-informed neural networks. The local adaptation of…

Machine Learning · Computer Science 2021-04-28 Ameya D. Jagtap , Kenji Kawaguchi , George Em Karniadakis

The classical Universal Approximation Theorem holds for neural networks of arbitrary width and bounded depth. Here we consider the natural `dual' scenario for networks of bounded width and arbitrary depth. Precisely, let $n$ be the number…

Machine Learning · Computer Science 2020-06-09 Patrick Kidger , Terry Lyons

Recently there has been much interest in understanding why deep neural networks are preferred to shallow networks. We show that, for a large class of piecewise smooth functions, the number of neurons needed by a shallow network to…

Machine Learning · Computer Science 2017-03-07 Shiyu Liang , R. Srikant

We study the expressivity of deep neural networks. Measuring a network's complexity by its number of connections or by its number of neurons, we consider the class of functions for which the error of best approximation with networks of a…

Functional Analysis · Mathematics 2020-07-20 Rémi Gribonval , Gitta Kutyniok , Morten Nielsen , Felix Voigtlaender

We propose a novel extension to symmetrized neural network operators by incorporating fractional and mixed activation functions. This study addresses the limitations of existing models in approximating higher-order smooth functions,…

Machine Learning · Statistics 2025-01-22 Rômulo Damasclin Chaves dos Santos , Jorge Henrique de Oliveira Sales

\citet{farrell2021deep} establish non-asymptotic high-probability bounds for general deep feedforward neural network (with rectified linear unit activation function) estimators, with \citet[Theorem 1]{farrell2021deep} achieving a suboptimal…

Econometrics · Economics 2025-12-11 Zhaoji Tang

We introduce a class of fully-connected neural networks whose activation functions, rather than being pointwise, rescale feature vectors by a function depending only on their norm. We call such networks radial neural networks, extending…

Machine Learning · Computer Science 2023-02-17 Iordan Ganev , Twan van Laarhoven , Robin Walters

Research in computational deep learning has directed considerable efforts towards hardware-oriented optimisations for deep neural networks, via the simplification of the activation functions, or the quantization of both activations and…

Machine Learning · Computer Science 2020-11-04 Gian Paolo Leonardi , Matteo Spallanzani

We investigate properties of neural networks that use both ReLU and $x^2$ as activation functions and build upon previous results to show that both analytic functions and functions in Sobolev spaces can be approximated by such networks of…

Machine Learning · Computer Science 2023-01-31 Vincent P. H. Goverse , Jad Hamdan , Jared Tanner

The possibility of approximating a continuous function on a compact subset of the real line by a feedforward single hidden layer neural network with a sigmoidal activation function has been studied in many papers. Such networks can…

Neural and Evolutionary Computing · Computer Science 2016-06-29 Namig J. Guliyev , Vugar E. Ismailov

We consider the approximation rates of shallow neural networks with respect to the variation norm. Upper bounds on these rates have been established for sigmoidal and ReLU activation functions, but it has remained an important open problem…

Machine Learning · Statistics 2021-09-10 Jonathan W. Siegel , Jinchao Xu

In this work, we examine the approximation capabilities of deep neural networks utilizing the Rectified Quadratic Unit (ReQU) activation function, defined as \(\max(0,x)^2\), for approximating H\"older-regular functions with respect to the…

Machine Learning · Computer Science 2024-11-12 Ahmed Abdeljawad

Recently, the authors of \cite{SYZ22} developed a neural network with width $36d(2d + 1)$ and depth $11$, which utilizes a special activation function called the elementary universal activation function, to achieve the super approximation…

Machine Learning · Computer Science 2025-06-17 Ayan Maiti , Michelle Michelle , Haizhao Yang

The scope of research in the domain of activation functions remains limited and centered around improving the ease of optimization or generalization quality of neural networks (NNs). However, to develop a deeper understanding of deep…

Machine Learning · Computer Science 2020-12-10 Mohit Goyal , Rajan Goyal , Brejesh Lall

In this paper, we prove that in the overparametrized regime, deep neural network provide universal approximations and can interpolate any data set, as long as the activation function is locally in $L^1(\RR)$ and not an affine function.…

Machine Learning · Computer Science 2024-04-26 Vlad-Raul Constantinescu , Ionel Popescu

In this paper we present a class of convolutional neural networks (CNNs) called non-overlapping CNNs in the study of approximation capabilities of CNNs. We prove that such networks with sigmoidal activation function are capable of…

Machine Learning · Computer Science 2022-09-28 Weike Chang

Some properties of generalized convexity for sets and for functions are identified in case of the reliability polynomials of two dual minimal networks. A method of approximating the reliability polynomials of two dual minimal network is…

Discrete Mathematics · Computer Science 2021-12-14 Gabriela Cristescu , Vlad-Florin Dragoi , Sorin-Horatiu Hoara

In this paper, we explain the universal approximation capabilities of deep residual neural networks through geometric nonlinear control. Inspired by recent work establishing links between residual networks and control systems, we provide a…

Machine Learning · Computer Science 2024-02-12 Paulo Tabuada , Bahman Gharesifard

We consider an existing conjecture addressing the asymptotic behavior of neural networks in the large width limit. The results that follow from this conjecture include tight bounds on the behavior of wide networks during stochastic gradient…

Machine Learning · Computer Science 2020-06-15 Kyle Aitken , Guy Gur-Ari

It has been shown that gradient descent can yield the zero training loss in the over-parametrized regime (the width of the neural networks is much larger than the number of data points). In this work, combining the ideas of some existing…

Optimization and Control · Mathematics 2019-11-05 Lei Li