ICML 2019oral121 citations
On Connected Sublevel Sets in Deep Learning
Abstract
This paper shows that every sublevel set of the loss function of a class of deep over-parameterized neural nets with piecewise linear activation functions is connected and unbounded. This implies that the loss has no bad local valleys and all of its global minima are connected within a unique and potentially very large global valley.
BibTeX
@InProceedings{pmlr-v97-nguyen19a,
title = {On Connected Sublevel Sets in Deep Learning},
author = {Nguyen, Quynh},
booktitle = {Proceedings of the 36th International Conference on Machine Learning},
pages = {4790--4799},
year = {2019},
editor = {Chaudhuri, Kamalika and Salakhutdinov, Ruslan},
volume = {97},
series = {Proceedings of Machine Learning Research},
month = {09--15 Jun},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v97/nguyen19a/nguyen19a.pdf},
url = {https://proceedings.mlr.press/v97/nguyen19a.html},
abstract = {This paper shows that every sublevel set of the loss function of a class of deep over-parameterized neural nets with piecewise linear activation functions is connected and unbounded. This implies that the loss has no bad local valleys and all of its global minima are connected within a unique and potentially very large global valley.}
}