Exploring Generalization in Deep Learning
Abstract
With a goal of understanding what drives generalization in deep networks, we consider several recently suggested explanations, including norm-based control, sharpness and robustness. We study how these measures can ensure generalization, highlighting the importance of scale normalization, and making a connection between sharpness and PAC-Bayes theory. We then investigate how well the measures explain different observed phenomena.
Cite
Text
Neyshabur et al. "Exploring Generalization in Deep Learning." Neural Information Processing Systems, 2017.Markdown
[Neyshabur et al. "Exploring Generalization in Deep Learning." Neural Information Processing Systems, 2017.](https://mlanthology.org/neurips/2017/neyshabur2017neurips-exploring/)BibTeX
@inproceedings{neyshabur2017neurips-exploring,
title = {{Exploring Generalization in Deep Learning}},
author = {Neyshabur, Behnam and Bhojanapalli, Srinadh and Mcallester, David and Srebro, Nati},
booktitle = {Neural Information Processing Systems},
year = {2017},
pages = {5947-5956},
url = {https://mlanthology.org/neurips/2017/neyshabur2017neurips-exploring/}
}