Activity Number:

127
 SPEED: Statistical Learning and Data Science Speed Session 1, Part 1

Type:

Contributed

Date/Time:

Monday, July 29, 2019 : 8:30 AM to 10:20 AM

Sponsor:

Section on Statistical Learning and Data Science

Abstract #302974

Presentation

Title:

Deep Learning and MARS: a Connection

Author(s):

Sophie Langer* and Michael Kohler and Adam Krzyzak

Companies:

Technische Universitaet Darmstadt and Technische Universitaet Darmstadt and Concordia University

Keywords:

Curse of dimensionality;
deep neural networks;
MARS;
nonparametric regression;
rate of convergence ;
piecewise partitioning

Abstract:

We consider least squares regression estimates using deep neural networks. We show that these estimates satisfy an oracle inequality, which implies that (up to a logarithmic factor) the error of these estimates is at least as small as the optimal possible error bound which one would expect for MARS in case that this procedure would work in the optimal way. As a result we show that our neural networks are able to achieve a dimensionality reduction in case that the regression function locally has low dimensionality. This assumption seems to be realistic in realworld applications, since selected highdimensional data are often confined to locallylowdimensional distributions. In our simulation study we provide numerical experiments to support our theoretical results and to compare our estimate with other conventional nonparametric regression estimates, especially with MARS.
