ML Anthology

RankMI: A Mutual Information Maximizing Ranking Loss

Mete Kemertas, Leila Pishdad, Konstantinos G. Derpanis, Afsaneh Fazly
CVPR 2020
doi:10.1109/CVPR42600.2020.01437 /cvpr/2020/kemertas2020cvpr-rankmi/

Abstract

We introduce an information-theoretic loss function, RankMI, and an associated training algorithm for deep representation learning for image retrieval. Our proposed framework consists of alternating updates to a network that estimates the divergence between distance distributions of matching and non-matching pairs of learned embeddings, and an embedding network that maximizes this estimate via sampled negatives. In addition, under this information-theoretic lens we draw connections between RankMI and commonly-used ranking losses, e.g., triplet loss. We extensively evaluate RankMI on several standard image retrieval datasets, namely, CUB-200-2011, CARS-196, and Stanford Online Products. Our method achieves competitive results or significant improvements over previous reported results on all datasets.

PDF CVPR Semantic Scholar

Cite

Text

Kemertas et al. "RankMI: A Mutual Information Maximizing Ranking Loss." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020. doi:10.1109/CVPR42600.2020.01437

Markdown

[Kemertas et al. "RankMI: A Mutual Information Maximizing Ranking Loss." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020.](https://mlanthology.org/cvpr/2020/kemertas2020cvpr-rankmi/) doi:10.1109/CVPR42600.2020.01437

BibTeX

@inproceedings{kemertas2020cvpr-rankmi,
  title     = {{RankMI: A Mutual Information Maximizing Ranking Loss}},
  author    = {Kemertas, Mete and Pishdad, Leila and Derpanis, Konstantinos G. and Fazly, Afsaneh},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2020},
  doi       = {10.1109/CVPR42600.2020.01437},
  url       = {https://mlanthology.org/cvpr/2020/kemertas2020cvpr-rankmi/}
}