Approximation of BV functions by neural networks: A regularity theory approach

12/15/2020
by   Benny Avelin, et al.
0

In this paper we are concerned with the approximation of functions by single hidden layer neural networks with ReLU activation functions on the unit circle. In particular, we are interested in the case when the number of data-points exceeds the number of nodes. We first study the convergence to equilibrium of the stochastic gradient flow associated with the cost function with a quadratic penalization. Specifically, we prove a Poincaré inequality for a penalized version of the cost function with explicit constants that are independent of the data and of the number of nodes. As our penalization biases the weights to be bounded, this leads us to study how well a network with bounded weights can approximate a given function of bounded variation (BV). Our main contribution concerning approximation of BV functions, is a result which we call the localization theorem. Specifically, it states that the expected error of the constrained problem, where the length of the weights are less than R, is of order R^-1/9 with respect to the unconstrained problem (the global optimum). The proof is novel in this topic and is inspired by techniques from regularity theory of elliptic partial differential equations. Finally we quantify the expected value of the global optimum by proving a quantitative version of the universal approximation theorem.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
05/13/2019

Towards a regularity theory for ReLU networks -- chain rule and global error estimates

Although for neural networks with locally Lipschitz continuous activatio...
research
01/29/2019

Approximation of functions by neural networks

We study the approximation of measurable functions on the hypercube by f...
research
03/29/2023

Optimal approximation of C^k-functions using shallow complex-valued neural networks

We prove a quantitative result for the approximation of functions of reg...
research
06/30/2020

Approximation Rates for Neural Networks with Encodable Weights in Smoothness Spaces

We examine the necessary and sufficient complexity of neural networks to...
research
02/26/2018

A representer theorem for deep neural networks

We propose to optimize the activation functions of a deep neural network...
research
02/09/2022

On the stability of unevenly spaced samples for interpolation and quadrature

Unevenly spaced samples from a periodic function are common in signal pr...
research
10/02/2019

The option pricing model based on time values: an application of the universal approximation theory on unbounded domains

Hutchinson, Lo and Poggio raised the question that if learning works can...

Please sign up or login with your details

Forgot password? Click here to reset