Digital Library[ Search Result ]
Learning with Noisy Labels using Sample Selection based on Language-Image Pre-trained Model
Bonggeon Cha, Minjin Choi, Jongwuk Lee
http://doi.org/10.5626/JOK.2023.50.6.511
Deep neural networks have significantly degraded generalization performance when learning with noisy labels. To address this problem, previous studies observed that the model learns clean samples first in the early learning stage, and based on this, sample selection methods that selectively train data by considering small-loss samples as clean samples have been used to improve performance. However, when noisy labels are similar to their ground truth(e.g., seal vs. otter), sample selection is not effective because the model learns noisy data in the early learning stage. In this paper, we propose a Sample selection with Language-Image Pre-trained model (SLIP) which effectively distinguishes and learns clean samples without the early learning stage by leveraging zero-shot predictions from a pre-trained language-image model. Our proposed method shows up to 18.45%p improved performance over previously proposed methods on CIFAR-10, CIFAR-100, and WebVision.
Improving Low Resource Chest X-ray Classification Accuracy by Combining Data Augmentation and Weakly Supervised Learning
http://doi.org/10.5626/JOK.2021.48.9.1027
Deep learning-based medical image analysis technology has been developed to the extent that it shows an accuracy surpassing the ability of a human radiologist. However, labeling sample data for use in learning medical images requires human experts and a great deal of time and expense. In addition, the training data for medical images has an unbalanced data distribution in many cases. For example, in the case of the ChestX-ray14 dataset, the difference between the number of data for infiltration and hernia is about 87 times. In this study, we proposed a method that combines the data augmentation algorithm Mixup and weakly supervised learning to improve the performance of data-imbalanced chest X-ray classification. The proposed method is to apply Mixup algorithm to a small number of labeled data and a large number of unlabeled data to alleviate data imbalance and performs curriculum learning that effectively utilizes the unlabeled data while cycling through the teacher model and the student model. Experimental results in an environment with a small number of labeled data and a large number of unlabeled data that can be considered in the medical field showed that the classification performance was improved by combining data augmentation and weakly supervised learning and that the cyclic curriculum learning was effective.
Search
Journal of KIISE
- ISSN : 2383-630X(Print)
- ISSN : 2383-6296(Electronic)
- KCI Accredited Journal
Editorial Office
- Tel. +82-2-588-9240
- Fax. +82-2-521-1352
- E-mail. chwoo@kiise.or.kr