Parametric Contrastive Learning

By Jiequan Cui, Zhisheng Zhong, Shu Liu, Bei Yu, Jiaya Jia. IEEE International Conference on Computer Vision, 2021

397 citations recorded by the source corpus.

In this paper, we propose Parametric Contrastive Learning (PaCo) to tackle long-tailed recognition. Based on theoretical analysis, we observe supervised contrastive loss tends to bias on high-frequency classes and thus increases the difficulty of imbalanced learning. We introduce a set of parametric class-wise learnable centers to rebalance from an optimization perspective. Further, we analyze our PaCo loss under a balanced setting. Our analysis demonstrates that PaCo can adaptively enhance the intensity of pushing samples of the same class close as more samples are pulled together with their corresponding centers and benefit hard example learning. Experiments on long-tailed CIFAR, ImageNet, Places, and iNaturalist 2018 manifest the new state-of-the-art for long-tailed recognition. On full ImageNet, models trained with PaCo loss surpass supervised contrastive learning across various ResNet backbones, e.g., our ResNet-200 achieves 81.8% top-1 accuracy. Our code is available at https://github.com/dvlab-research/Parametric-Contrastive-Learning.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/6900eaf98c9d35aaf2f1f8a411ca62343fe449a2

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Jiashi Feng
2. Shuicheng Yan
3. Yifan Zhang
4. Bingyi Kang
5. Bryan Hooi
6. Yunchao Wei
7. Yi Yang
8. Dacheng Tao

In Computer Vision and Pattern Recognition

1. Jiashi Feng
2. Shuicheng Yan
3. Yifan Zhang
4. Bingyi Kang
5. Luc Van Gool
6. Yi Yang
7. Dacheng Tao
8. Bryan Hooi

In Radiology, Nuclear Medicine and Imaging

1. Yifan Zhang
2. Jiashi Feng
3. Shuicheng Yan
4. Bingyi Kang

Citations in the corpus

• Deep Long-Tailed Learning: A Survey (2021, methodology): “Moreover, when handling real long-tailed applications, PaCo [121] is recommended to use thanks to its promising performance and open-source code.”