English

Positive Difference Distribution for Image Outlier Detection using Normalizing Flows and Contrastive Data

Machine Learning 2023-04-28 v2

Abstract

Detecting test data deviating from training data is a central problem for safe and robust machine learning. Likelihoods learned by a generative model, e.g., a normalizing flow via standard log-likelihood training, perform poorly as an outlier score. We propose to use an unlabelled auxiliary dataset and a probabilistic outlier score for outlier detection. We use a self-supervised feature extractor trained on the auxiliary dataset and train a normalizing flow on the extracted features by maximizing the likelihood on in-distribution data and minimizing the likelihood on the contrastive dataset. We show that this is equivalent to learning the normalized positive difference between the in-distribution and the contrastive feature density. We conduct experiments on benchmark datasets and compare to the likelihood, the likelihood ratio and state-of-the-art anomaly detection methods.

Keywords

Cite

@article{arxiv.2208.14024,
  title  = {Positive Difference Distribution for Image Outlier Detection using Normalizing Flows and Contrastive Data},
  author = {Robert Schmier and Ullrich Köthe and Christoph-Nikolas Straehle},
  journal= {arXiv preprint arXiv:2208.14024},
  year   = {2023}
}
R2 v1 2026-06-25T02:04:45.749Z