Risk Bounds for Multi-layer Perceptrons through Spectra of Integral Operators

02/28/2020
by   Meyer Scetbon, et al.
0

We characterize the behavior of integral operators associated with multi-layer perceptrons in two eigenvalue decay regimes. We obtain as a result sharper risk bounds for multi-layer perceptrons highlighting their behavior in high dimensions. Doing so, we also improve on previous results on integral operators related to power series kernels on spheres, with sharper eigenvalue decay estimates in a wider range of eigenvalue decay regimes.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/01/2021

Computation of transmission eigenvalues by the regularized Schur complement for the boundary integral operators

This paper is devoted to the computation of transmission eigenvalues in ...
research
01/27/2020

Solving Maxwell's Eigenvalue Problem via Isogeometric Boundary Elements and a Contour Integral Method

We solve Maxwell's eigenvalue problem via isogeometric boundary elements...
research
05/01/2022

On the speed of uniform convergence in Mercer's theorem

The classical Mercer's theorem claims that a continuous positive definit...
research
10/22/2021

Refined decay bounds on the entries of spectral projectors associated with sparse Hermitian matrices

Spectral projectors of Hermitian matrices play a key role in many applic...
research
02/10/2023

K-Spectral Sets

We use results in [M. Crouzeix and A. Greenbaum,Spectral sets: numerical...
research
02/13/2022

Information Density in Multi-Layer Resistive Memories

Resistive memories store information in a crossbar arrangement of two-te...

Please sign up or login with your details

Forgot password? Click here to reset