Exploiting Mixed Unlabeled Data for Detecting Samples of Seen and Unseen Out-of-Distribution Classes
This addresses the high labeling cost in OOD detection for real-world applications by leveraging mixed unlabeled data, though it is incremental as it builds on existing OOD detection methods.
The paper tackles the problem of Out-of-Distribution (OOD) detection with limited labeled data and abundant mixed unlabeled data, proposing the Adaptive In-Out-aware Learning (AIOL) method that adaptively selects samples and uses entropy and data augmentation to improve performance, achieving superior results on benchmark datasets.
Out-of-Distribution (OOD) detection is essential in real-world applications, which has attracted increasing attention in recent years. However, most existing OOD detection methods require many labeled In-Distribution (ID) data, causing a heavy labeling cost. In this paper, we focus on the more realistic scenario, where limited labeled data and abundant unlabeled data are available, and these unlabeled data are mixed with ID and OOD samples. We propose the Adaptive In-Out-aware Learning (AIOL) method, in which we employ the appropriate temperature to adaptively select potential ID and OOD samples from the mixed unlabeled data and consider the entropy over them for OOD detection. Moreover, since the test data in realistic applications may contain OOD samples whose classes are not in the mixed unlabeled data (we call them unseen OOD classes), data augmentation techniques are brought into the method to further improve the performance. The experiments are conducted on various benchmark datasets, which demonstrate the superiority of our method.