Learning non-linear invariants for unsupervised out-of-distribution detection

Doorenbos, Lars; Sznitman, Raphael; Márquez-Neila, Pablo (October 2024). Learning non-linear invariants for unsupervised out-of-distribution detection. In: European Conference on Computer Vision.

[img] Text
2407.04022v1.pdf - Published Version
Restricted to registered users only
Available under License Publisher holds Copyright.

Download (2MB)

The inability of deep learning models to handle data drawn from unseen distributions has sparked much interest in unsupervised out-of-distribution (U-OOD) detection, as it is crucial for reliable deep learning models. Despite considerable attention, theoretically-motivated approaches are few and far between, with most methods building on top of some form of heuristic. Recently, U-OOD was formalized in the context of data invariants, allowing a clearer understanding of how to characterize U-OOD, and methods leveraging affine invariants have attained state-of-the-art results on large-scale benchmarks. Nevertheless, the restriction to affine invariants hinders the expressiveness of the approach. In this work, we broaden the affine invariants formulation to a more general case and propose a framework consisting of a normalizing flow-like architecture capable of learning non-linear invariants. Our novel approach achieves state-of-the-art results on an extensive U-OOD benchmark, and we demonstrate its further applicability to tabular data. Finally, we show our method has the same desirable properties as those based on affine invariants.

Item Type:

Conference or Workshop Item (Paper)

Division/Institute:

10 Strategic Research Centers > ARTORG Center for Biomedical Engineering Research
10 Strategic Research Centers > ARTORG Center for Biomedical Engineering Research > ARTORG Center - AI in Medical Imaging Laboratory

Graduate School:

Graduate School for Cellular and Biomedical Sciences (GCB)

UniBE Contributor:

Doorenbos, Lars Jelte, Sznitman, Raphael, Márquez Neila, Pablo

Subjects:

500 Science > 570 Life sciences; biology
600 Technology > 610 Medicine & health
000 Computer science, knowledge & systems
600 Technology

Language:

English

Submitter:

Lars Jelte Doorenbos

Date Deposited:

16 Jul 2024 11:38

Last Modified:

16 Jul 2024 11:38

ArXiv ID:

2407.04022v1

BORIS DOI:

10.48350/199035

URI:

https://boris.unibe.ch/id/eprint/199035

Actions (login required)

Edit item Edit item
Provide Feedback