Out-Of-Distribution Detection with Diversification (Provably)
This work addresses the reliability of machine learning models in deployment by improving OOD detection, but it is incremental as it builds on existing methods using auxiliary outliers.
The paper tackles the problem of out-of-distribution (OOD) detection by addressing the limited diversity in auxiliary outlier data used during training, which hinders generalization to unknown OOD data. It proposes a method called diverseMix that enhances diversity efficiently, achieving superior performance on benchmarks.
Out-of-distribution (OOD) detection is crucial for ensuring reliable deployment of machine learning models. Recent advancements focus on utilizing easily accessible auxiliary outliers (e.g., data from the web or other datasets) in training. However, we experimentally reveal that these methods still struggle to generalize their detection capabilities to unknown OOD data, due to the limited diversity of the auxiliary outliers collected. Therefore, we thoroughly examine this problem from the generalization perspective and demonstrate that a more diverse set of auxiliary outliers is essential for enhancing the detection capabilities. However, in practice, it is difficult and costly to collect sufficiently diverse auxiliary outlier data. Therefore, we propose a simple yet practical approach with a theoretical guarantee, termed Diversity-induced Mixup for OOD detection (diverseMix), which enhances the diversity of auxiliary outlier set for training in an efficient way. Extensive experiments show that diverseMix achieves superior performance on commonly used and recent challenging large-scale benchmarks, which further confirm the importance of the diversity of auxiliary outliers.