Hypercone Assisted Contour Generation for Out-of-Distribution Detection

Annita Vapsi, Andrés Muñoz Garza, Nancy Thomas, Keshav Ramani, Daniel Borrajo

Keywords: OOD detection, Out-of-distribution detection, Computer Vision, Deep Learning, Representation Learning

TL;DR: HACk-OOD is a novel, training-agnostic OOD detection method that constructs hypercones in the feature space to approximate in-distribution contours of classes, without making distributional assumptions or explicitly training for OOD detection.

Abstract: Recent advances in the field of out-of-distribution (OOD) detection have placed great emphasis on learning better representations suited to this task. While there have been distance-based approaches, distributional awareness has seldom been exploited for better performance. We present HACk-OOD, a novel OOD detection method that makes no distributional assumption about the data, but automatically adapts to its distribution. Specifically, HACk-OOD constructs a set of hypercones by maximizing the angular distance to neighbors in a given data-point's vicinity, to approximate the contour within which in-distribution (ID) data-points lie. Experimental results show state-of-the-art FPR@95 and AUROC performance on Near-OOD detection and on Far-OOD detection on the challenging CIFAR-100 benchmark without explicitly training for OOD performance.

Primary Area: other topics in machine learning (i.e., none of the above)

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 11290