ReweightOOD: Loss Reweighting for Distance-based OOD Detection

Sudarshan Regmi, Bibek Panthi, Yifei Ming, Prashnna Kumar Gyawali, Danail Stoyanov, Binod Bhattarai

Primary Area: societal considerations including fairness, safety, privacy

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.

Keywords: Out of Distribution Detection

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

TL;DR: We propose loss reweighting to obtain suitable embedding for OOD detection.

Abstract: Out-of-Distribution (OOD) detection is crucial for ensuring the safety and reliability of neural networks in critical applications. Distance-based OOD detection is based on the assumption that OOD samples are mapped far from In-Distribution (ID) clusters in embedding space. A recent approach for obtaining OOD-detection-friendly embedding space has been contrastive optimization of pulling similar pairs and pushing apart dissimilar pairs. It assigns equal significance to all similarity instances with the implicit objective of maximizing the mean proximity between samples with their corresponding hypothetical class centroids. However, the emphasis should be directed towards reducing the Minimum Enclosing Sphere (MES) for each class and achieving higher inter-class dispersion to effectively mitigate the potential for ID-OOD overlap. Optimizing low-signal dissimilar pairs might potentially act against achieving maximal inter-class dispersion while less-optimized similar pairs prevent achieving smaller MES. Based on this, we propose a reweighting scheme \textbf{ReweightOOD}, that adopts the similarity optimization which prioritizes the optimization of less-optimized contrasting pairs while assigning lower importance to already well-optimized contrasting pairs. Such a reweighting scheme serves to minimize the MES for each class while achieving maximal inter-class dispersion. Experimental results on a challenging CIFAR100 benchmark using ResNet-18 network demonstrate that the proposed reweighting scheme improves the FPR metric by a whopping ~38\% in comparison to the baseline. In various classification datasets, our method outperforms existing methods, making it a promising solution for enhancing OOD detection capabilities in neural networks.

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: 226