Representation formulas and pointwise properties for Barron functions

06/10/2020
by   Weinan E, et al.
0

We study the natural function space for infinitely wide two-layer neural networks and establish different representation formulae. In two cases, we describe the space explicitly up to isomorphism. Using a convenient representation, we study the pointwise properties of two-layer networks and show that functions whose singular set is fractal or curved (for example distance functions from smooth submanifolds) cannot be represented by infinitely wide two-layer networks with finite path-norm.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
07/30/2020

On the Banach spaces associated with multi-layer ReLU networks: Function representation, approximation theory and gradient descent dynamics

We develop Banach spaces for ReLU neural networks of finite depth L and ...
research
07/31/2020

The Kolmogorov-Arnold representation theorem revisited

There is a longstanding debate whether the Kolmogorov-Arnold representat...
research
04/15/2020

A function space analysis of finite neural networks with insights from sampling theory

This work suggests using sampling theory to analyze the function space r...
research
03/15/2021

Representation Theorem for Matrix Product States

In this work, we investigate the universal representation capacity of th...
research
11/20/2020

A global universality of two-layer neural networks with ReLU activations

In the present study, we investigate a universality of neural networks, ...
research
05/21/2020

Kolmogorov Width Decay and Poor Approximators in Machine Learning: Shallow Neural Networks, Random Feature Models and Neural Tangent Kernels

We establish a scale separation of Kolmogorov width type between subspac...
research
03/26/2021

Some properties of the parking function poset

In 1980, Edelman defined a poset on objects called the noncrossing 2-par...

Please sign up or login with your details

Forgot password? Click here to reset