A Simple Fix to Mahalanobis Distance for Improving Near-OOD Detection

Abstract

Mahalanobis distance (MD) is a simple and popular post-processing method for detecting out-of-distribution (OOD) inputs in neural networks. We analyze its failure modes for near-OOD detection and propose a simple fix called relative Mahalanobis distance (RMD) which improves performance and is more robust to hyperparameter choice. On a wide selection of challenging vision, language, and biology OOD benchmarks (CIFAR-100 vs CIFAR-10, CLINC OOD intent detection, Genomics OOD), we show that RMD meaningfully improves upon MD performance (by up to 15% AUROC on genomics OOD).

References in corpus (4)

• Learning Transferable Visual Models From Natural Language Supervision
• Exploring the Limits of Out-of-Distribution Detection
• Contrastive Training for Improved Out-of-Distribution Detection
• Why is the Mahalanobis Distance Effective for Anomaly Detection?

Cited by in corpus (10)

• Generalized Out-of-Distribution Detection: A Survey
• A Unified Survey on Anomaly, Novelty, Open-Set, and Out-of-Distribution Detection: Solutions and Future Challenges
• No True State-of-the-Art? OOD Detection Methods are Inconsistent across Datasets
• ScatterUQ: Interactive Uncertainty Visualizations for Multiclass Deep Learning Problems
• Distance Matters For Improving Performance Estimation Under Covariate Shift
• A Distance-based Anomaly Detection Framework for Deep Reinforcement Learning
• Rainproof: An Umbrella To Shield Text Generators From Out-Of-Distribution Data
• Decomposing Representations for Deterministic Uncertainty Estimation
• Unsupervised Out-of-Distribution Dialect Detection with Mahalanobis Distance
• FROB: Few-shot ROBust Model for Classification and Out-of-Distribution Detection