Label-Noise Learning with Intrinsically Long-Tailed Data

作     者：Lu, Yang Zhang, Yiliang Han, Bo Cheung, Yiu-Ming Wang, Hanzi 

作者机构：Fujian Key Laboratory of Sensing and Computing for Smart City School of Informatics Xiamen University Xiamen China Key Laboratory of Multimedia Trusted Perception and Efficient Computing Ministry of Education of China Xiamen University Xiamen China Department of Computer Science Hong Kong Baptist University Hong Kong 

出 版 物：《arXiv》 (arXiv)

年 卷 期：2022年

核心收录：

主　　题：Deep learning 

摘      要：Label noise is one of the key factors that lead to the poor generalization of deep learning models. Existing label-noise learning methods usually assume that the ground-truth classes of the training data are balanced. However, the real-world data is often imbalanced, leading to the inconsistency between observed and intrinsic class distribution with label noises. In this case, it is hard to distinguish clean samples from noisy samples on the intrinsic tail classes with the unknown intrinsic class distribution. In this paper, we propose a learning framework for label-noise learning with intrinsically long-tailed data. Specifically, we propose two-stage bi-dimensional sample selection (TABASCO) to better separate clean samples from noisy samples, especially for the tail classes. TABASCO consists of two new separation metrics that complement each other to compensate for the limitation of using a single metric in sample separation. Extensive experiments on benchmarks demonstrate the effectiveness of our method. Our code is available at https://***/Wakings/TABASCO. Copyright © 2022, The Authors. All rights reserved.