Searching for Accurate Binary Neural Architectures

Mingzhu Shen, Kai Han, Chunjing Xu, Yunhe Wang

ICCVW 2019 pp. 2041-2044

doi:10.1109/ICCVW.2019.00256 /iccvw/2019/shen2019iccvw-searching/

Abstract

Binary neural networks have attracted tremendous attention due to the efficiency for deploying them on mobile devices. Since the weak expression ability of binary weights and features, their accuracy is usually much lower than that of full-precision (i.e. 32-bit) models. Here we present a new frame work for automatically searching for compact but accurate binary neural networks. In practice, number of channels in each layer will be encoded into the search space and optimized using the evolutionary algorithm. Experiments conducted on benchmark datasets and neural architectures demonstrate that our searched binary networks can achieve the performance of full-precision models with acceptable increments on model sizes and calculations.

PDF ICCVW Semantic Scholar

Cite

Text

Shen et al. "Searching for Accurate Binary Neural Architectures." IEEE/CVF International Conference on Computer Vision Workshops, 2019. doi:10.1109/ICCVW.2019.00256

Markdown

[Shen et al. "Searching for Accurate Binary Neural Architectures." IEEE/CVF International Conference on Computer Vision Workshops, 2019.](https://mlanthology.org/iccvw/2019/shen2019iccvw-searching/) doi:10.1109/ICCVW.2019.00256

BibTeX

@inproceedings{shen2019iccvw-searching,
  title     = {{Searching for Accurate Binary Neural Architectures}},
  author    = {Shen, Mingzhu and Han, Kai and Xu, Chunjing and Wang, Yunhe},
  booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
  year      = {2019},
  pages     = {2041-2044},
  doi       = {10.1109/ICCVW.2019.00256},
  url       = {https://mlanthology.org/iccvw/2019/shen2019iccvw-searching/}
}