Optical Instruments, Volume. 47, Issue 2, 50(2025)
Image classification approach using AllMix for label noise learning
Figures & Tables(4)
Table 1. Experimental results of AllMix with different levels of symmetric noise on the CIFAR10 and CIFAR100 datasets
View tableView in ArticleTable 1. Experimental results of AllMix with different levels of symmetric noise on the CIFAR10 and CIFAR100 datasets
模型 CIFAR10 CIFAR100 20% 50% 80% 90% 20% 50% 80% 90% Mixup[ 15 ]Best 95.6 87.1 71.6 52.2 67.8 57.3 30.8 14.6 Last 92.3 77.6 46.7 43.9 66.0 46.6 17.6 8.1 Co-teaching+[ 18 ]Best 89.5 85.7 67.4 47.9 65.6 51.8 27.9 13.7 Last 88.2 84.1 45.5 30.1 64.1 45.3 15.5 8.8 P-correction[ 19 ]Best 92.4 89.1 77.5 58.9 69.4 57.5 31.1 15.3 Last 92.0 88.7 76.5 58.2 68.1 56.4 20.7 8.8 Meta-Learning[ 20 ]Best 92.9 89.3 77.4 58.7 68.5 59.2 42.4 19.5 Last 92.0 88.8 76.1 58.3 67.7 58.0 40.1 14.3 M-correction[ 21 ]Best 94.0 92.0 86.8 69.1 73.9 66.1 48.2 24.3 Last 93.8 91.9 86.6 68.7 73.4 65.4 47.6 20.5 DivideMix[ 12 ]Best 96.1 94.6 93.2 76.0 77.3 74.6 60.2 31.5 Last 95.7 94.4 92.9 75.4 76.9 74.2 59.6 31.0 CSS[ 22 ]Best 96.5 96.3 95.6 94.4 79.1 77.7 72.5 68.7 Last Ours Best 96.1 95.3 93.9 81.0 76.3 73.7 63.0 41.6 Last 95.9 95.1 93.6 80.4 75.7 72.0 62.7 41.1
Table 2. Results of the AllMix ablation experiments on CIFAR10 dataset
View tableView in ArticleTable 2. Results of the AllMix ablation experiments on CIFAR10 dataset
AllMix 模型 20% 50% 80% 90% AllMix Best 96.1 95.3 93.9 81.0 Last 95.9 95.1 93.6 80.4 AllMix w/o $ {L}_{U2} $ Best 96.1 93.2 88.9 72.9 Last 95.9 70.4 88.5 72.3 AllMix w/o $ {L}_{{\mathrm{contrast}}} $ Best 95.8 94.7 90.2 77.5 Last 95.6 94.5 90.0 77.1
Tools
Get Citation
Copy Citation Text
Yemin QIU, Rongfu ZHANG, Chen HE, Ziye YANG, Guyu GAO. Image classification approach using AllMix for label noise learning[J]. Optical Instruments, 2025, 47(2): 50
Paper Information
Category:
Received: Feb. 27, 2024
Accepted: --
Published Online: May. 30, 2025
The Author Email: Rongfu ZHANG (zrf@usst.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20