How Low Can You Go? Surfacing Prototypical In-Distribution Samples for Unsupervised Anomaly Detection
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2023
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866910671696822272 |
|---|---|
| author | Meissen, Felix Getzner, Johannes Ziller, Alexander Turgut, Özgün Kaissis, Georgios Menten, Martin J. Rueckert, Daniel |
| author_facet | Meissen, Felix Getzner, Johannes Ziller, Alexander Turgut, Özgün Kaissis, Georgios Menten, Martin J. Rueckert, Daniel |
| contents | Unsupervised anomaly detection (UAD) alleviates large labeling efforts by training exclusively on unlabeled in-distribution data and detecting outliers as anomalies. Generally, the assumption prevails that large training datasets allow the training of higher-performing UAD models. However, in this work, we show that UAD with extremely few training samples can already match -- and in some cases even surpass -- the performance of training with the whole training dataset. Building upon this finding, we propose an unsupervised method to reliably identify prototypical samples to further boost UAD performance. We demonstrate the utility of our method on seven different established UAD benchmarks from computer vision, industrial defect detection, and medicine. With just 25 selected samples, we even exceed the performance of full training in $25/67$ categories in these benchmarks. Additionally, we show that the prototypical in-distribution samples identified by our proposed method generalize well across models and datasets and that observing their sample selection criteria allows for a successful manual selection of small subsets of high-performing samples. Our code is available at https://anonymous.4open.science/r/uad_prototypical_samples/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2312_03804 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | How Low Can You Go? Surfacing Prototypical In-Distribution Samples for Unsupervised Anomaly Detection Meissen, Felix Getzner, Johannes Ziller, Alexander Turgut, Özgün Kaissis, Georgios Menten, Martin J. Rueckert, Daniel Computer Vision and Pattern Recognition Unsupervised anomaly detection (UAD) alleviates large labeling efforts by training exclusively on unlabeled in-distribution data and detecting outliers as anomalies. Generally, the assumption prevails that large training datasets allow the training of higher-performing UAD models. However, in this work, we show that UAD with extremely few training samples can already match -- and in some cases even surpass -- the performance of training with the whole training dataset. Building upon this finding, we propose an unsupervised method to reliably identify prototypical samples to further boost UAD performance. We demonstrate the utility of our method on seven different established UAD benchmarks from computer vision, industrial defect detection, and medicine. With just 25 selected samples, we even exceed the performance of full training in $25/67$ categories in these benchmarks. Additionally, we show that the prototypical in-distribution samples identified by our proposed method generalize well across models and datasets and that observing their sample selection criteria allows for a successful manual selection of small subsets of high-performing samples. Our code is available at https://anonymous.4open.science/r/uad_prototypical_samples/ |
| title | How Low Can You Go? Surfacing Prototypical In-Distribution Samples for Unsupervised Anomaly Detection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2312.03804 |