Detecting Out-of-Distribution Inputs in Deep Neural Networks Using an Early-Layer Output

TLDR

Deep neural networks excel at image classification yet often misclassify out‑of‑distribution inputs, making reliable OOD detection an ongoing challenge. This study proposes a new OOD detection method that can be applied to any existing classifier without requiring OOD samples. The method trains a one‑class classifier on the output of an early layer of the original network using only its training data. Across several low‑ and high‑dimensional datasets, the approach outperforms state‑of‑the‑art detectors on multiple metrics.

Abstract

Deep neural networks achieve superior performance in challenging tasks such as image classification. However, deep classifiers tend to incorrectly classify out-of-distribution (OOD) inputs, which are inputs that do not belong to the classifier training distribution. Several approaches have been proposed to detect OOD inputs, but the detection task is still an ongoing challenge. In this paper, we propose a new OOD detection approach that can be easily applied to an existing classifier and does not need to have access to OOD samples. The detector is a one-class classifier trained on the output of an early layer of the original classifier fed with its original training set. We apply our approach to several low- and high-dimensional datasets and compare it to the state-of-the-art detection approaches. Our approach achieves substantially better results over multiple metrics.