NeurIPS 2018poster44 citations
TETRIS: TilE-matching the TRemendous Irregular Sparsity
Yu Ji, Ling Liang, Lei Deng, Youyang Zhang, Youhui Zhang, Yuan Xie
Abstract
Compressing neural networks by pruning weights with small magnitudes can significantly reduce the computation and storage cost. Although pruning makes the model smaller, it is difficult to get practical speedup in modern computing platforms such as CPU and GPU due to the irregularity. Structural pruning has attract a lot of research interest to make sparsity hardware-friendly. Increasing the sparsity granularity can lead to better hardware utilization, but it will compromise the sparsity for maintaining accuracy.
BibTeX
@inproceedings{NEURIPS2018_89885ff2,
author = {Ji, Yu and Liang, Ling and Deng, Lei and Zhang, Youyang and Zhang, Youhui and Xie, Yuan},
booktitle = {Advances in Neural Information Processing Systems},
editor = {S. Bengio and H. Wallach and H. Larochelle and K. Grauman and N. Cesa-Bianchi and R. Garnett},
pages = {},
publisher = {Curran Associates, Inc.},
title = {TETRIS: TilE-matching the TRemendous Irregular Sparsity},
url = {https://proceedings.neurips.cc/paper_files/paper/2018/file/89885ff2c83a10305ee08bd507c1049c-Paper.pdf},
volume = {31},
year = {2018}
}