Abstract

Deep classifiers have achieved great success in visual recognition. However, real-world data is long-tailed by nature, leading to the mismatch between training and testing distributions. In this paper, we show that Softmax function, though used in most classification tasks, gives a biased gradient estimation under the long-tailed setup. This paper presents Balanced Softmax, an elegant unbiased extension of Softmax, to accommodate the label distribution shift between training and testing. Theoretically, we derive the generalization bound for multiclass Softmax regression and show our loss minimizes the bound. In addition, we introduce Balanced Meta-Softmax, applying a complementary Meta Sampler to estimate the optimal class sample rate and further improve long-tailed learning. In our experiments, we demonstrate that Balanced Meta-Softmax outperforms state-of-the-art long-tailed classification solutions on both visual recognition and instance segmentation tasks.

Paper

Jiawei Ren, Cunjun Yu, Shunan Sheng, Xiao Ma, Haiyu Zhao, Shuai Yi, Hongsheng Li

Balanced Meta-Softmax for Long-Tailed Visual Recognition

NeurIPS 2020, [PDF], [Code]

BibTex

@inproceedings{

Ren2020BALMS,

title={Balanced Meta-Softmax for Long-Tailed Visual Recognition},

author={Jiawei Ren and Cunjun Yu and Shunan Sheng and Xiao Ma and Haiyu Zhao and Shuai Yi and Hongsheng Li},

booktitle={Advances in Neural Information Processing Systems},

month = {December},

year={2020}

}