SEN: A Novel Feature Normalization Dissimilarity Measure for Prototypical Few-Shot Learning Networks
Abstract
In this paper, we equip Prototypical Networks (PNs) with a novel dissimilarity measure to enable discriminative feature normalization for few-shot learning. The embedding onto the hypersphere requires no direct normalization and is easy to optimize. Our theoretical analysis shows that the proposed dissimilarity measure, denoted the Squared root of the Euclidean distance and the Norm distance (SEN), forces embedding points to be attracted to its correct prototype, while being repelled from all other prototypes, keeping the norm of all points the same. The resulting SEN PN outperforms the regular PN with a considerable margin, with no additional parameters as well as with negligible computational overhead.
Cite
Text
Nguyen et al. "SEN: A Novel Feature Normalization Dissimilarity Measure for Prototypical Few-Shot Learning Networks." Proceedings of the European Conference on Computer Vision (ECCV), 2020. doi:10.1007/978-3-030-58592-1_8Markdown
[Nguyen et al. "SEN: A Novel Feature Normalization Dissimilarity Measure for Prototypical Few-Shot Learning Networks." Proceedings of the European Conference on Computer Vision (ECCV), 2020.](https://mlanthology.org/eccv/2020/nguyen2020eccv-sen/) doi:10.1007/978-3-030-58592-1_8BibTeX
@inproceedings{nguyen2020eccv-sen,
title = {{SEN: A Novel Feature Normalization Dissimilarity Measure for Prototypical Few-Shot Learning Networks}},
author = {Nguyen, Van Nhan and Løkse, Sigurd and Wickstrøm, Kristoffer and Kampffmeyer, Michael and Roverso, Davide and Jenssen, Robert},
booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)},
year = {2020},
doi = {10.1007/978-3-030-58592-1_8},
url = {https://mlanthology.org/eccv/2020/nguyen2020eccv-sen/}
}