Distribution-Balanced Loss for Multi-Label Classification in Long-Tailed Datasets

07/19/2020

∙

by Tong Wu, et al.

∙

Tsinghua University

∙

The Chinese University of Hong Kong

∙

We present a new loss function called Distribution-Balanced Loss for the multi-label recognition problems that exhibit long-tailed class distributions. Compared to conventional single-label classification problem, multi-label recognition problems are often more challenging due to two significant issues, namely the co-occurrence of labels and the dominance of negative labels (when treated as multiple binary classification problems). The Distribution-Balanced Loss tackles these issues through two key modifications to the standard binary cross-entropy loss: 1) a new way to re-balance the weights that takes into account the impact caused by label co-occurrence, and 2) a negative tolerant regularization to mitigate the over-suppression of negative labels. Experiments on both Pascal VOC and COCO show that the models trained with this new loss function achieve significant performance gains over existing methods. Code and models are available at: https://github.com/wutong16/DistributionBalancedLoss .

READ FULL TEXT

Distribution-Balanced Loss for Multi-Label Classification in Long-Tailed Datasets

Balancing Methods for Multi-label Text Classification with Long-Tailed Class Distribution

Robust Asymmetric Loss for Multi-Label Long-Tailed Learning

Understanding Gender and Racial Disparities in Image Recognition Models

Using Ontologies To Improve Performance In Massively Multi-label Prediction Models

Combining Metric Learning and Attention Heads For Accurate and Efficient Multilabel Image Classification

Caption supervision enables robust learners

Asymmetric Polynomial Loss For Multi-Label Classification

Code Repositories