Learning in High Dimension Always Amounts to Extrapolation

This page cannot be shown here. You can still read it on the original site — the toolbar below keeps your place in the directory.

In this post I’ll attempt to shed some light on the conclusion that is drawn (in part 1 ) from the above image: We shouldn’t use interpolation/extrapolation in the way the terms are defined below when talking about generalization, because for high dimensional data deep learning models always have to extrapolate, regardless of the dimension of the underlying data manifold. The image is taken from…

Read on /2021/11/06/extra.html ↗