SSD: A Unified Framework for Self-Supervised Outlier Detection

TL;DR

This paper introduces SSD, a self-supervised outlier detection framework that effectively identifies out-of-distribution samples using only unlabeled in-distribution data, outperforming existing methods and matching supervised approaches.

Contribution

The paper presents SSD, a novel self-supervised outlier detection method that leverages Mahalanobis distance in feature space, achieving state-of-the-art results without requiring labeled data.

Findings

SSD outperforms most unlabeled data detectors significantly.

SSD matches or exceeds the performance of supervised detectors.

Extensions to few-shot and labeled data scenarios further improve detection accuracy.

Abstract

We ask the following question: what training information is required to design an effective outlier/out-of-distribution (OOD) detector, i.e., detecting samples that lie far away from the training distribution? Since unlabeled data is easily accessible for many applications, the most compelling approach is to develop detectors based on only unlabeled in-distribution data. However, we observe that most existing detectors based on unlabeled data perform poorly, often equivalent to a random prediction. In contrast, existing state-of-the-art OOD detectors achieve impressive performance but require access to fine-grained data labels for supervised training. We propose SSD, an outlier detector based on only unlabeled in-distribution data. We use self-supervised representation learning followed by a Mahalanobis distance based detection in the feature space. We demonstrate that SSD outperforms most existing detectors based on unlabeled data by a large margin. Additionally, SSD even achieves performance on par, and sometimes even better, with supervised training based detectors. Finally, we expand our detection framework with two key extensions. First, we formulate few-shot OOD detection, in which the detector has access to only one to five samples from each class of the targeted OOD dataset. Second, we extend our framework to incorporate training data labels, if available. We find that our novel detection framework based on SSD displays enhanced performance with these extensions, and achieves state-of-the-art performance. Our code is publicly available at https://github.com/inspire-group/SSD.