English

Anomaly Detection Requires Better Representations

Machine Learning 2022-10-20 v1 Computer Vision and Pattern Recognition

Abstract

Anomaly detection seeks to identify unusual phenomena, a central task in science and industry. The task is inherently unsupervised as anomalies are unexpected and unknown during training. Recent advances in self-supervised representation learning have directly driven improvements in anomaly detection. In this position paper, we first explain how self-supervised representations can be easily used to achieve state-of-the-art performance in commonly reported anomaly detection benchmarks. We then argue that tackling the next generation of anomaly detection tasks requires new technical and conceptual improvements in representation learning.

Keywords

Cite

@article{arxiv.2210.10773,
  title  = {Anomaly Detection Requires Better Representations},
  author = {Tal Reiss and Niv Cohen and Eliahu Horwitz and Ron Abutbul and Yedid Hoshen},
  journal= {arXiv preprint arXiv:2210.10773},
  year   = {2022}
}

Comments

Accepted to ECCV SSLWIN Workshop (2022)

R2 v1 2026-06-28T04:01:31.469Z