RSG: A Simple but Effective Module for Learning Imbalanced Datasets
Abstract
Imbalanced datasets widely exist in practice and area great challenge for training deep neural models with agood generalization on infrequent classes. In this work, wepropose a new rare-class sample generator (RSG) to solvethis problem. RSG aims to generate some new samplesfor rare classes during training, and it has in particularthe following advantages: (1) it is convenient to use andhighly versatile, because it can be easily integrated intoany kind of convolutional neural network, and it works wellwhen combined with different loss functions, and (2) it isonly used during the training phase, and therefore, no ad-ditional burden is imposed on deep neural networks duringthe testing phase. In extensive experimental evaluations, weverify the effectiveness of RSG. Furthermore, by leveragingRSG, we obtain competitive results on Imbalanced CIFARand new state-of-the-art results on Places-LT, ImageNet-LT, and iNaturalist 2018. The source code is available at https://github.com/Jianf-Wang/RSG.
Cite
@article{arxiv.2106.09859,
title = {RSG: A Simple but Effective Module for Learning Imbalanced Datasets},
author = {Jianfeng Wang and Thomas Lukasiewicz and Xiaolin Hu and Jianfei Cai and Zhenghua Xu},
journal= {arXiv preprint arXiv:2106.09859},
year = {2021}
}
Comments
To appear at CVPR 2021. We propose a flexible data generation/data augmentation module for long-tailed classification. Codes are available at: https://github.com/Jianf-Wang/RSG