Efficient Multi-Domain Learning by Covariance Normalization
Abstract
The problem of multi-domain learning of deep networks is considered. An adaptive layer is induced per target domain and a novel procedure, denoted covariance normalization (CovNorm), proposed to reduce its parameters. CovNorm is a data driven method of fairly simple implementation, requiring two principal component analyzes (PCA) and fine-tuning of a mini-adaptation layer. Nevertheless, it is shown, both theoretically and experimentally, to have several advantages over previous approaches, such as batch normalization or geometric matrix approximations. Furthermore, CovNorm can be deployed both when target datasets are available sequentially or simultaneously. Experiments show that, in both cases, it has performance comparable to a fully fine-tuned network, using as few as 0.13% of the corresponding parameters per target domain.
Cite
Text
Li and Vasconcelos. "Efficient Multi-Domain Learning by Covariance Normalization." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019. doi:10.1109/CVPR.2019.00557Markdown
[Li and Vasconcelos. "Efficient Multi-Domain Learning by Covariance Normalization." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019.](https://mlanthology.org/cvpr/2019/li2019cvpr-efficient/) doi:10.1109/CVPR.2019.00557BibTeX
@inproceedings{li2019cvpr-efficient,
title = {{Efficient Multi-Domain Learning by Covariance Normalization}},
author = {Li, Yunsheng and Vasconcelos, Nuno},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
year = {2019},
doi = {10.1109/CVPR.2019.00557},
url = {https://mlanthology.org/cvpr/2019/li2019cvpr-efficient/}
}