Deep Double Descent: Where Bigger Models and More Data Hurt

2020 inproceedings nakkiran2020doubledescent ○ Not yet verified

Authors

Preetum Nakkiran, Gal Kaplun, Yamini Bansal, Tristan Yang, Boaz Barak, Ilya Sutskever

Venue

ICLR 2020

Abstract

Demonstrates that double descent occurs across model size, training epochs, and dataset size in modern deep networks. Introduces effective model complexity to unify these phenomena and shows regimes where more data hurts.

BibTeX

Local Entry

@inproceedings{nakkiran2020doubledescent,
  title = {Deep Double Descent: Where Bigger Models and More Data Hurt},
  author = {Preetum Nakkiran and Gal Kaplun and Yamini Bansal and Tristan Yang and Boaz Barak and Ilya Sutskever},
  year = {2020},
  booktitle = {ICLR 2020},
  url = {https://arxiv.org/abs/1912.02292},
  abstract = {Demonstrates that double descent occurs across model size, training epochs, and dataset size in modern deep networks. Introduces effective model complexity to unify these phenomena and shows regimes where more data hurts.}
}

From OPENALEX

@inproceedings{nakkiran2020doubledescent,
  title = {Deep double descent: where bigger models and more data hurt*},
  author = {Preetum Nakkiran and Gal Kaplun and Yamini Bansal and Tristan Yang and Boaz Barak and Ilya Sutskever},
  year = {2021},
  booktitle = {Journal of Statistical Mechanics Theory and Experiment},
  doi = {10.1088/1742-5468/ac3a74}
}