Deep Double Descent: Where Bigger Models and More Data HurtDownload PDF

25 Sept 2019, 19:23 (modified: 10 Feb 2022, 11:42)ICLR 2020 Conference Blind SubmissionReaders: Everyone
Original Pdf: pdf
Code: [![Papers with Code](/images/pwc_icon.svg) 3 community implementations](
Data: [CIFAR-10](, [CIFAR-100](
TL;DR: We demonstrate, and characterize, realistic settings where bigger models are worse, and more data hurts.
Abstract: 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.
Keywords: deep learning, double descent, optimization, SGD, complexity
10 Replies