Self-Augmentation Based on Noise-Robust Probabilistic Model For Noisy Labelsopen access
- Authors
- Park, B.W.; Park, S.W.; Kwon, Junseok
- Issue Date
- Nov-2022
- Publisher
- Institute of Electrical and Electronics Engineers Inc.
- Keywords
- Noise measurement; Probabilistic logic; Mixture models; Gaussian distribution; Neural networks; Noise robustness; Computer vision; deep learning; probability distribution
- Citation
- IEEE Access, v.10, pp 116141 - 116151
- Pages
- 11
- Journal Title
- IEEE Access
- Volume
- 10
- Start Page
- 116141
- End Page
- 116151
- URI
- https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/59527
- DOI
- 10.1109/ACCESS.2022.3219810
- ISSN
- 2169-3536
- Abstract
- Learning deep neural networks from noisy labels is challenging, because high-capacity networks attempt to describe data even with noisy class labels. In this study, we propose a self-augmentation method without additional parameters, which handles noisy labeled data based on small-loss criteria. To this end, we use small-loss samples by introducing a noise-robust probabilistic model based on a Gaussian mixture model (GMM), in which small-loss samples follow class-conditional Gaussian distributions. With this sample augmentation using the GMM-based probabilistic model, we can effectively solve over-parameterization problems induced by label inconsistency in small-loss samples. We further enhance the quality of the small-loss samples using our data-adaptive selection strategy. Consequently, our method prevents networks from over-parameterization and enhances their generalization performance. Experimental results demonstrate that our method outperforms state-of-the-art methods for learning with noisy labels on several benchmark datasets. The proposed method produced a remarkable performance gap of up to 12% compared with the previous state-of-the-art methods on CIFAR dataset. Author
- Files in This Item
-
- Appears in
Collections - College of Software > School of Computer Science and Engineering > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.