NeurIPS 2020poster28 citations

Sharp Representation Theorems for ReLU Networks with Precise Dependence on Depth

Guy Bresler, Dheeraj Nagaraj

Abstract

We prove dimension free representation results for neural networks with D ReLU layers under square loss for a class of functions G_D defined in the paper. These results capture the precise benefits of depth in the following sense:

BibTeX
@inproceedings{NEURIPS2020_78f7d96e,
 author = {Bresler, Guy and Nagaraj, Dheeraj},
 booktitle = {Advances in Neural Information Processing Systems},
 editor = {H. Larochelle and M. Ranzato and R. Hadsell and M.F. Balcan and H. Lin},
 pages = {10697--10706},
 publisher = {Curran Associates, Inc.},
 title = {Sharp Representation Theorems for ReLU Networks with Precise Dependence on Depth},
 url = {https://proceedings.neurips.cc/paper_files/paper/2020/file/78f7d96ea21ccae89a7b581295f34135-Paper.pdf},
 volume = {33},
 year = {2020}
}