Calibrated Out-of-Distribution Detection with a Generic Representation
This addresses safety-critical deployment issues for vision models, though it appears incremental as it builds on existing representation learning approaches.
The paper tackles out-of-distribution detection in vision models by proposing GROOD, a method that leverages generic pre-trained representations and formulates detection as a Neyman-Pearson task with calibrated scores. It achieves state-of-the-art performance on multiple benchmarks, reaching near-perfect results on several.
Out-of-distribution detection is a common issue in deploying vision models in practice and solving it is an essential building block in safety critical applications. Most of the existing OOD detection solutions focus on improving the OOD robustness of a classification model trained exclusively on in-distribution (ID) data. In this work, we take a different approach and propose to leverage generic pre-trained representation. We propose a novel OOD method, called GROOD, that formulates the OOD detection as a Neyman-Pearson task with well calibrated scores and which achieves excellent performance, predicated by the use of a good generic representation. Only a trivial training process is required for adapting GROOD to a particular problem. The method is simple, general, efficient, calibrated and with only a few hyper-parameters. The method achieves state-of-the-art performance on a number of OOD benchmarks, reaching near perfect performance on several of them. The source code is available at https://github.com/vojirt/GROOD.