期刊
PATTERN RECOGNITION
卷 130, 期 -, 页码 -出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.patcog.2022.108777
关键词
Dynamic mutual training; Inter-model disagreement; Noisy pseudo label; Semi-supervised learning
资金
- National Key Research and Development Program of China [2019YFC1521104]
- National Natural Science Foundation of China [72192821, 61972157]
- Shanghai Municipal Science and Technology Major Project [2021SHZDZX0102]
- Shanghai Science and Technology Commission [21511101200]
- Art major project of National Social Science Fund [I8ZD22]
- High-Level Talent Program for Innovation and Entrepreneurship (ShuangChuang Doctor) of Jiangsu Province [JSSCBS20211220]
- Wu Wenjun Honorary Doctoral Scholarship, AI Institute, Shanghai Jiao Tong University
Recent semi-supervised learning methods have used pseudo supervision, particularly self-training methods, which generate unreliable pseudo labels. This paper proposes a new approach called Dynamic Mutual Training (DMT) that leverages inter-model disagreement to locate pseudo label errors and achieve state-of-the-art performance in image classification and semantic segmentation.
Recent semi-supervised learning methods use pseudo supervision as core idea, especially self-training methods that generate pseudo labels. However, pseudo labels are unreliable. Self-training methods usually rely on single model prediction confidence to filter low-confidence pseudo labels, thus remaining high-confidence errors and wasting many low-confidence correct labels. In this paper, we point out it is difficult for a model to counter its own errors. Instead, leveraging inter-model disagreement between different models is a key to locate pseudo label errors. With this new viewpoint, we propose mutual training between two different models by a dynamically re-weighted loss function, called Dynamic Mutual Training (DMT). We quantify inter-model disagreement by comparing predictions from two different models to dynamically re-weight loss in training, where a larger disagreement indicates a possible error and corresponds to a lower loss value. Extensive experiments show that DMT achieves state-of-the-art performance in both image classification and semantic segmentation. Our codes are released at https:///github.com/voldemortX/DST-CBC. (C) 2022 Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据