A Systematic Comparison of Training Objectives for Out-of-Distribution Detection in Image Classification
This work provides insights into the impact of different training objectives on OOD detection performance for researchers and practitioners in safety-sensitive AI applications.
This paper systematically compares four training objectives for out-of-distribution (OOD) detection in image classification. They found that Cross-Entropy Loss, Prototype Loss, and AP Loss achieve comparable in-distribution accuracy, with Cross-Entropy Loss providing the most consistent OOD performance.
Out-of-distribution (OOD) detection is critical in safety-sensitive applications. While this challenge has been addressed from various perspectives, the influence of training objectives on OOD behavior remains comparatively underexplored. In this paper, we present a systematic comparison of four widely used training objectives: Cross-Entropy Loss, Prototype Loss, Triplet Loss, and Average Precision (AP) Loss, spanning probabilistic, prototype-based, metric-learning, and ranking-based supervision, for OOD detection in image classification under standardized OpenOOD protocols. Across CIFAR-10/100 and ImageNet-200, we find that Cross-Entropy Loss, Prototype Loss, and AP Loss achieve comparable in-distribution accuracy, while Cross-Entropy Loss provides the most consistent near- and far-OOD performance overall; the other objectives can be competitive in specific settings.