Automated Multi-Stage Compression of Neural Networks

Julia Gusak, Maksym Kholyavchenko, Evgeny Ponomarev, Larisa Markeeva, Philip Blagoveschensky, Andrzej Cichocki, Ivan V. Oseledets

ICCVW 2019 pp. 2501-2508

doi:10.1109/ICCVW.2019.00306 /iccvw/2019/gusak2019iccvw-automated/

Abstract

Low-rank tensor approximations are very promising for compression of deep neural networks. We propose a new simple and efficient iterative approach, which alternates low-rank factorization with smart rank selection and fine-tuning. We demonstrate the efficiency of our method comparing to non-iterative ones. Our approach improves the compression rate while maintaining the accuracy for a variety of tasks.

ICCVW Semantic Scholar

Cite

Text

Gusak et al. "Automated Multi-Stage Compression of Neural Networks." IEEE/CVF International Conference on Computer Vision Workshops, 2019. doi:10.1109/ICCVW.2019.00306

Markdown

[Gusak et al. "Automated Multi-Stage Compression of Neural Networks." IEEE/CVF International Conference on Computer Vision Workshops, 2019.](https://mlanthology.org/iccvw/2019/gusak2019iccvw-automated/) doi:10.1109/ICCVW.2019.00306

BibTeX

@inproceedings{gusak2019iccvw-automated,
  title     = {{Automated Multi-Stage Compression of Neural Networks}},
  author    = {Gusak, Julia and Kholyavchenko, Maksym and Ponomarev, Evgeny and Markeeva, Larisa and Blagoveschensky, Philip and Cichocki, Andrzej and Oseledets, Ivan V.},
  booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
  year      = {2019},
  pages     = {2501-2508},
  doi       = {10.1109/ICCVW.2019.00306},
  url       = {https://mlanthology.org/iccvw/2019/gusak2019iccvw-automated/}
}