Addressing Feature Suppression in Unsupervised Visual Representations

Tianhong Li, Lijie Fan, Yuan Yuan, Hao He, Yonglong Tian, Rogerio Feris, Piotr Indyk, Dina Katabi; Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2023, pp. 1411-1420

Abstract


Contrastive learning is one of the fastest growing research areas in machine learning due to its ability to learn useful representations without labeled data. However, contrastive learning is susceptible to feature suppression - i.e., it may discard important information relevant to the task of interest, and learn irrelevant features. Past work has addressed this limitation via handcrafted data augmentations that eliminate irrelevant information. This approach however does not work across all datasets and tasks. Further, data augmentations fail in addressing feature suppression in multi-attribute classification when one attribute can suppress features relevant to other attributes. In this paper, we analyze the objective function of contrastive learning and formally prove that it is vulnerable to feature suppression. We then present Predictive Contrastive Learning (PrCL), a framework for learning unsupervised representations that are robust to feature suppression. The key idea is to force the learned representation to predict the input, and hence prevent it from discarding important information. Extensive experiments verify that PrCL is robust to feature suppression and outperforms state-of-the-art contrastive learning methods on a variety of datasets and tasks.

Related Material


[pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Li_2023_WACV, author = {Li, Tianhong and Fan, Lijie and Yuan, Yuan and He, Hao and Tian, Yonglong and Feris, Rogerio and Indyk, Piotr and Katabi, Dina}, title = {Addressing Feature Suppression in Unsupervised Visual Representations}, booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)}, month = {January}, year = {2023}, pages = {1411-1420} }