Feature Density Estimation for Out-of-Distribution Detection via Normalizing Flows
This addresses the problem of safe deployment of learning systems in open-world settings for AI practitioners, but it is incremental as it builds on existing normalizing flow methods.
The paper tackles out-of-distribution (OOD) detection in image classification by using feature density estimation with normalizing flows, achieving 98.2% AUROC for ImageNet-1k vs. Textures, which exceeds state-of-the-art by 7.8%.
Out-of-distribution (OOD) detection is a critical task for safe deployment of learning systems in the open world setting. In this work, we investigate the use of feature density estimation via normalizing flows for OOD detection and present a fully unsupervised approach which requires no exposure to OOD data, avoiding researcher bias in OOD sample selection. This is a post-hoc method which can be applied to any pretrained model, and involves training a lightweight auxiliary normalizing flow model to perform the out-of-distribution detection via density thresholding. Experiments on OOD detection in image classification show strong results for far-OOD data detection with only a single epoch of flow training, including 98.2% AUROC for ImageNet-1k vs. Textures, which exceeds the state of the art by 7.8%. We additionally explore the connection between the feature space distribution of the pretrained model and the performance of our method. Finally, we provide insights into training pitfalls that have plagued normalizing flows for use in OOD detection.