Heatmap-based Out-of-Distribution Detection
This addresses the problem of detecting anomalous inputs in neural networks for researchers and practitioners, but it appears incremental as it builds on existing OOD detection methods with a new visualization-focused approach.
The paper tackles out-of-distribution (OOD) detection by framing it as a neural network explanation problem, using heatmaps to visualize in- and out-of-distribution regions, and reports that their approach mostly outperforms prior work on fixed classifiers trained on datasets like CIFAR-10, CIFAR-100, and Tiny ImageNet.
Our work investigates out-of-distribution (OOD) detection as a neural network output explanation problem. We learn a heatmap representation for detecting OOD images while visualizing in- and out-of-distribution image regions at the same time. Given a trained and fixed classifier, we train a decoder neural network to produce heatmaps with zero response for in-distribution samples and high response heatmaps for OOD samples, based on the classifier features and the class prediction. Our main innovation lies in the heatmap definition for an OOD sample, as the normalized difference from the closest in-distribution sample. The heatmap serves as a margin to distinguish between in- and out-of-distribution samples. Our approach generates the heatmaps not only for OOD detection, but also to indicate in- and out-of-distribution regions of the input image. In our evaluations, our approach mostly outperforms the prior work on fixed classifiers, trained on CIFAR-10, CIFAR-100 and Tiny ImageNet. The code is publicly available at: https://github.com/jhornauer/heatmap_ood.