TY - JOUR
AU - Sutskever, Ilya
AB - 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.
TI - Deep double descent: where bigger models and more data hurt*               *This article is an updated version of: Nakkiran P, Kaplun G, Bansal Y, Yang T, Barak B and Sutskever I 2020 Deep double descent: where bigger models and more data hurt Int. Conf. Learning Representations.
JF - The Planetary Science Journal
DO - 10.1088/1742-5468/ac3a74
DA - 2021-12-01
UR - https://www.deepdyve.com/lp/iop-publishing/deep-double-descent-where-bigger-models-and-more-data-hurt-this-5fy1ZpmKLi
VL - 2021
IS - 12
DP - DeepDyve
ER -