Dissecting Mahalanobis: How Feature Geometry and Normalization Shape OOD Detection
This addresses the reliability of OOD detection for deploying deep learning models, but it is incremental as it builds on existing Mahalanobis-based methods.
The paper tackled the problem of understanding how feature geometry and normalization affect Mahalanobis distance methods for out-of-distribution (OOD) detection in deep learning, finding that these methods aren't universally reliable and proposing a radially scaled normalization method that significantly improves OOD detection performance.
Out-of-distribution (OOD) detection is critical for the reliable deployment of deep learning models. hile Mahalanobis distance methods are widely used, the impact of representation geometry and normalization on their performance is not fully understood, which may limit their downstream application. To address this gap, we conducted a comprehensive empirical study across diverse image foundation models, datasets, and distance normalization schemes. First, our analysis shows that Mahalanobis-based methods aren't universally reliable. Second, we define the ideal geometry for data representations and demonstrate that spectral and intrinsic-dimensionality metrics can accurately predict a model's OOD performance. Finally, we analyze how normalization impacts OOD performance. Building upon these studies, we propose radially scaled $\ell_2$ normalization, a method that generalizes the standard $\ell_2$ normalization recently applied to Mahalanobis-based OOD detection. Our approach introduces a tunable parameter to directly control the radial geometry of the feature space, systematically contracting or expanding representations to significantly improve OOD detection performance. By bridging the gap between representation geometry, normalization, and OOD performance, our findings offer new insights into the design of more effective and reliable deep learning models.