Loading…

Deep double descent: where bigger models and more data hurt

We show that a variety of modern deep learning tasks exhibit a ‘double-descent’ phenomenon where, as we increase model size, performance first gets worse and then gets better. Moreover, we show that double descent occurs not just as a function of model size, but also as a function of the number of t...

Full description

Saved in:
Bibliographic Details
Published in:Journal of statistical mechanics 2021-12, Vol.2021 (12), p.124003
Main Authors: Nakkiran, Preetum, Kaplun, Gal, Bansal, Yamini, Yang, Tristan, Barak, Boaz, Sutskever, Ilya
Format: Article
Language:English
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We show that a variety of modern deep learning tasks exhibit a ‘double-descent’ phenomenon where, as we increase model size, performance first gets worse and then gets better. Moreover, we show that double descent occurs not just as a function of model size, but also as a function of the number of training epochs. We unify the above phenomena by defining a new complexity measure we call the effective model complexity and conjecture a generalized double descent with respect to this measure. Furthermore, our notion of model complexity allows us to identify certain regimes where increasing (even quadrupling) the number of train samples actually hurts test performance.
ISSN:1742-5468
1742-5468
DOI:10.1088/1742-5468/ac3a74