Self-Tuning Self-Supervised Anomaly Detection

Jaemin Yoo, Lingxiao Zhao, Leman Akoglu

Primary Area: unsupervised, self-supervised, semi-supervised, and supervised representation learning

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Keywords: Anomaly detection, Self-supervised learning, Data augmentation, Hyperparameter optimization

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

TL;DR: We propose the first framework for end-to-end hyperparameter optimization for self-supervised anomaly detection.

Abstract: Self-supervised learning (SSL) has emerged as a promising paradigm that presents supervisory signals to real-world problems, bypassing the extensive cost of manual labeling. Consequently, self-supervised anomaly detection (SSAD) has seen a recent surge of interest, since SSL is especially attractive for unsupervised tasks. However, recent works have reported that the choice of a data augmentation function has significant impact on the accuracy of SSAD, posing augmentation search as an essential but nontrivial problem with the lack of labeled validation data. In this paper, we introduce ST-SSAD (Self-Tuning Self-Supervised Anomaly Detection), the first systematic approach for rigorous augmentation tuning on SSAD. To this end, our work presents two key contributions. The first is a new unsupervised validation loss that quantifies the alignment between the augmented training data and the (unlabeled) test data. Second, we present new differentiable augmentation functions, allowing data augmentation hyperparameter(s) to be tuned end-to-end via our proposed validation loss. Experiments on two testbeds with semantic class anomalies and subtle industrial defects show that a systematic tuning of augmentation gives significant performance gains over current practices. All our code and testbeds are available at https://anonymous.4open.science/r/ST-SSAD.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 2588