2024-04-18-ahmadian24a.md

title	software	abstract	layout	series	publisher	issn	id	month	tex_title	firstpage	lastpage	page	order	cycles	bibtex_author	author	date	address	container-title	volume	genre	issued	pdf	extras
Unsupervised Novelty Detection in Pretrained Representation Space with Locally Adapted Likelihood Ratio	https://github.com/aahmadian-liu/pretrained-novelty-loclr	Detecting novelties given unlabeled examples of normal data is a challenging task in machine learning, particularly when the novel and normal categories are semantically close. Large deep models pretrained on massive datasets can provide a rich representation space in which the simple k-nearest neighbor distance works as a novelty measure. However, as we show in this paper, the basic k-NN method might be insufficient in this context due to ignoring the ’local geometry’ of the distribution over representations as well as the impact of irrelevant ’background features’. To address this, we propose a fully unsupervised novelty detection approach that integrates the flexibility of k-NN with a locally adapted scaling of dimensions based on the ’neighbors of nearest neighbor’ and computing a ’likelihood ratio’ in pretrained (self-supervised) representation spaces. Our experiments with image data show the advantage of this method when off-the-shelf vision transformers (e.g., pretrained by DINO) are used as the feature extractor without any fine-tuning.	inproceedings	Proceedings of Machine Learning Research	PMLR	2640-3498	ahmadian24a	0	Unsupervised Novelty Detection in Pretrained Representation Space with Locally Adapted Likelihood Ratio	874	882	874-882	874	false	Ahmadian, Amirhossein and Ding, Yifan and Eilertsen, Gabriel and Lindsten, Fredrik	given family Amirhossein Ahmadian given family Yifan Ding given family Gabriel Eilertsen given family Fredrik Lindsten	2024-04-18		Proceedings of The 27th International Conference on Artificial Intelligence and Statistics	238	inproceedings	date-parts 2024 4 18	https://proceedings.mlr.press/v238/ahmadian24a/ahmadian24a.pdf