期刊
ELECTRONICS
卷 10, 期 5, 页码 -出版社
MDPI
DOI: 10.3390/electronics10050567
关键词
out-of-distribution detection; confidence score; distance metric learning; Siamese network; triplet network; ensemble method; deep learning
资金
- Korea Institute of Science and Technology Information (KISTI)
The newly proposed OOD detection method uses an ensemble of features trained by a softmax-based classifier and a network based on distance metric learning (DML). Through complementary interaction, the trained feature space fits well on the Gaussian distribution by class, allowing for efficient detection of OOD data by setting a threshold in the trained feature space. The overall performance of the proposed approach is shown to be superior to other methods, including the state-of-the-art approach, on various combinations of datasets.
To design an efficient deep learning model that can be used in the real-world, it is important to detect out-of-distribution (OOD) data well. Various studies have been conducted to solve the OOD problem. The current state-of-the-art approach uses a confidence score based on the Mahalanobis distance in a feature space. Although it outperformed the previous approaches, the results were sensitive to the quality of the trained model and the dataset complexity. Herein, we propose a novel OOD detection method that can train more efficient feature space for OOD detection. The proposed method uses an ensemble of the features trained using the softmax-based classifier and the network based on distance metric learning (DML). Through the complementary interaction of these two networks, the trained feature space has a more clumped distribution and can fit well on the Gaussian distribution by class. Therefore, OOD data can be efficiently detected by setting a threshold in the trained feature space. To evaluate the proposed method, we applied our method to various combinations of image datasets. The results show that the overall performance of the proposed approach is superior to those of other methods, including the state-of-the-art approach, on any combination of datasets.
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