Mahalanobis++: Improving OOD Detection via Feature Normalization
This work improves reliability for safety-critical applications by enhancing OOD detection, though it is incremental as it builds on existing Mahalanobis distance methods.
The paper tackled the inconsistency in out-of-distribution (OOD) detection performance across models by addressing feature norm variations that violate Gaussian assumptions, and showed that simple L2-normalization significantly improves Mahalanobis distance-based methods, outperforming other recent approaches in extensive experiments on 44 models.
Detecting out-of-distribution (OOD) examples is an important task for deploying reliable machine learning models in safety-critial applications. While post-hoc methods based on the Mahalanobis distance applied to pre-logit features are among the most effective for ImageNet-scale OOD detection, their performance varies significantly across models. We connect this inconsistency to strong variations in feature norms, indicating severe violations of the Gaussian assumption underlying the Mahalanobis distance estimation. We show that simple $\ell_2$-normalization of the features mitigates this problem effectively, aligning better with the premise of normally distributed data with shared covariance matrix. Extensive experiments on 44 models across diverse architectures and pretraining schemes show that $\ell_2$-normalization improves the conventional Mahalanobis distance-based approaches significantly and consistently, and outperforms other recently proposed OOD detection methods.