ATOM: Robustifying Out-of-distribution Detection Using Outlier Mining

Abstract

Detecting out-of-distribution (OOD) inputs is critical for safely deploying deep learning models in an open-world setting. However, existing OOD detection solutions can be brittle in the open world, facing various types of adversarial OOD inputs. While methods leveraging auxiliary OOD data have emerged, our analysis on illuminative examples reveals a key insight that the majority of auxiliary OOD examples may not meaningfully improve or even hurt the decision boundary of the OOD detector, which is also observed in empirical results on real data. In this paper, we provide a theoretically motivated method, Adversarial Training with informative Outlier Mining (ATOM), which improves the robustness of OOD detection. We show that, by mining informative auxiliary OOD data, one can significantly improve OOD detection performance, and somewhat surprisingly, generalize to unseen adversarial attacks. ATOM achieves state-of-the-art performance under a broad family of classic and adversarial OOD evaluation tasks. For example, on the CIFAR-10 in-distribution dataset, ATOM reduces the FPR (at TPR 95%) by up to 57.99% under adversarial OOD inputs, surpassing the previous best baseline by a large margin.

arXiv commentPaper published at European Conference on Machine Learning (ECML'21)

References in corpus (14)

• A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks
• LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop
• A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks
• Deep Anomaly Detection with Outlier Exposure
• A Downsampled Variant of ImageNet as an Alternative to the CIFAR datasets
• TurkerGaze: Crowdsourcing Saliency with Webcam based Eye Tracking
• Are Labels Required for Improving Adversarial Robustness?
• Outlier Exposure with Confidence Control for Out-of-Distribution Detection
• Functional Adversarial Attacks
• Adversarially Robust Generalization Just Requires More Unlabeled Data
• Can Autonomous Vehicles Identify, Recover From, and Adapt to Distribution Shifts?
• Discriminative out-of-distribution detection for semantic segmentation
• Confidence estimation in Deep Neural networks via density modelling
• Certifiably Adversarially Robust Detection of Out-of-Distribution Data