NeurIPS 2019poster104 citations

Explaining Landscape Connectivity of Low-cost Solutions for Multilayer Nets

Rohith Kuditipudi, Xiang Wang, Holden Lee, Yi Zhang, Zhiyuan Li, Wei Hu, Rong Ge, Sanjeev Arora

Abstract

Mode connectivity is a surprising phenomenon in the loss landscape of deep nets. Optima---at least those discovered by gradient-based optimization---turn out to be connected by simple paths on which the loss function is almost constant. Often, these paths can be chosen to be piece-wise linear, with as few as two segments.

BibTeX
@inproceedings{NEURIPS2019_46a4378f,
 author = {Kuditipudi, Rohith and Wang, Xiang and Lee, Holden and Zhang, Yi and Li, Zhiyuan and Hu, Wei and Ge, Rong and Arora, Sanjeev},
 booktitle = {Advances in Neural Information Processing Systems},
 editor = {H. Wallach and H. Larochelle and A. Beygelzimer and F. d\textquotesingle Alch\'{e}-Buc and E. Fox and R. Garnett},
 pages = {},
 publisher = {Curran Associates, Inc.},
 title = {Explaining Landscape Connectivity of Low-cost Solutions for Multilayer Nets},
 url = {https://proceedings.neurips.cc/paper_files/paper/2019/file/46a4378f835dc8040c8057beb6a2da52-Paper.pdf},
 volume = {32},
 year = {2019}
}
Explaining Landscape Connectivity of Low-cost Solutions for Multilayer Nets · NeurIPS 2019