Improving Out-of-Distribution Detection via Dynamic Covariance Calibration

Kaiyu Guo, Zijian Wang, Tan Pan, Brian C. Lovell, Mahsa Baktashmotlagh
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:21067-21082, 2025.

Abstract

Out-of-Distribution (OOD) detection is essential for the trustworthiness of AI systems. Methods using prior information (i.e., subspace-based methods) have shown effective performance by extracting information geometry to detect OOD data with a more appropriate distance metric. However, these methods fail to address the geometry distorted by ill-distributed samples, due to the limitation of statically extracting information geometry from the training distribution. In this paper, we argue that the influence of ill-distributed samples can be corrected by dynamically adjusting the prior geometry in response to new data. Based on this insight, we propose a novel approach that dynamically updates the prior covariance matrix using real-time input features, refining its information. Specifically, we reduce the covariance along the direction of real-time input features and constrain adjustments to the residual space, thus preserving essential data characteristics and avoiding effects on unintended directions in the principal space. We evaluate our method on two pre-trained models for the CIFAR dataset and five pre-trained models for ImageNet-1k, including the self-supervised DINO model. Extensive experiments demonstrate that our approach significantly enhances OOD detection across various models. The code is released at https://github.com/workerbcd/ooddcc.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-guo25m, title = {Improving Out-of-Distribution Detection via Dynamic Covariance Calibration}, author = {Guo, Kaiyu and Wang, Zijian and Pan, Tan and Lovell, Brian C. and Baktashmotlagh, Mahsa}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {21067--21082}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/guo25m/guo25m.pdf}, url = {https://proceedings.mlr.press/v267/guo25m.html}, abstract = {Out-of-Distribution (OOD) detection is essential for the trustworthiness of AI systems. Methods using prior information (i.e., subspace-based methods) have shown effective performance by extracting information geometry to detect OOD data with a more appropriate distance metric. However, these methods fail to address the geometry distorted by ill-distributed samples, due to the limitation of statically extracting information geometry from the training distribution. In this paper, we argue that the influence of ill-distributed samples can be corrected by dynamically adjusting the prior geometry in response to new data. Based on this insight, we propose a novel approach that dynamically updates the prior covariance matrix using real-time input features, refining its information. Specifically, we reduce the covariance along the direction of real-time input features and constrain adjustments to the residual space, thus preserving essential data characteristics and avoiding effects on unintended directions in the principal space. We evaluate our method on two pre-trained models for the CIFAR dataset and five pre-trained models for ImageNet-1k, including the self-supervised DINO model. Extensive experiments demonstrate that our approach significantly enhances OOD detection across various models. The code is released at https://github.com/workerbcd/ooddcc.} }
Endnote
%0 Conference Paper %T Improving Out-of-Distribution Detection via Dynamic Covariance Calibration %A Kaiyu Guo %A Zijian Wang %A Tan Pan %A Brian C. Lovell %A Mahsa Baktashmotlagh %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-guo25m %I PMLR %P 21067--21082 %U https://proceedings.mlr.press/v267/guo25m.html %V 267 %X Out-of-Distribution (OOD) detection is essential for the trustworthiness of AI systems. Methods using prior information (i.e., subspace-based methods) have shown effective performance by extracting information geometry to detect OOD data with a more appropriate distance metric. However, these methods fail to address the geometry distorted by ill-distributed samples, due to the limitation of statically extracting information geometry from the training distribution. In this paper, we argue that the influence of ill-distributed samples can be corrected by dynamically adjusting the prior geometry in response to new data. Based on this insight, we propose a novel approach that dynamically updates the prior covariance matrix using real-time input features, refining its information. Specifically, we reduce the covariance along the direction of real-time input features and constrain adjustments to the residual space, thus preserving essential data characteristics and avoiding effects on unintended directions in the principal space. We evaluate our method on two pre-trained models for the CIFAR dataset and five pre-trained models for ImageNet-1k, including the self-supervised DINO model. Extensive experiments demonstrate that our approach significantly enhances OOD detection across various models. The code is released at https://github.com/workerbcd/ooddcc.
APA
Guo, K., Wang, Z., Pan, T., Lovell, B.C. & Baktashmotlagh, M.. (2025). Improving Out-of-Distribution Detection via Dynamic Covariance Calibration. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:21067-21082 Available from https://proceedings.mlr.press/v267/guo25m.html.

Related Material