paper-with-me

Papers

Certifiably Adversarially Robust Detection of Out-of-Distribution Data

2020-07-16 · NeurIPS 2020 12 · Julian Bitterwolf, Alexander Meinke, Matthias Hein

Deep neural networks are known to be overconfident when applied to out-of-distribution (OOD) inputs which clearly do not belong to any class. This is a problem in safety-critical applications since a reliable assessment of the uncertainty of a classifier is a key property, allowing the system to trigger human intervention or to transfer into a safe state. In this paper, we aim for certifiable worst case guarantees for OOD detection by enforcing not only low confidence at the OOD point but also in an $l_\infty$-ball around it. For this purpose, we use interval bound propagation (IBP) to upper bound the maximal confidence in the $l_\infty$-ball and minimize this upper bound during training time. We show that non-trivial bounds on the confidence for OOD data generalizing beyond the OOD dataset seen at training time are possible. Moreover, in contrast to certified adversarial robustness which typically comes with significant loss in prediction performance, certified guarantees for worst case OOD detection are possible without much loss in accuracy.

📄 PDF Abstract BibTeX arXiv:2007.08473

Code (2)

https://gitlab.com/Bitterwolf/GOOD 공식 구현 pytorch
j-cb/GOOD pytorch

Tasks

Adversarial RobustnessOut of Distribution (OOD) Detection

Similar Papers 제목 키워드 기반

Certified Robustness to Label-Flipping Attacks via Randomized Smoothing

2020-02-07 · ICML 2020 1 · Elan Rosenfeld, Ezra Winston, Pradeep Ravikumar, J. Zico Kolter

Machine learning algorithms are known to be susceptible to data poisoning attacks, where an adversary manipulates the training data to degrade performance of the resulting classifier. In this work, we present a unifying …

Data PoisoningGeneral ClassificationMulti-class Classification

Learning Better Certified Models from Empirically-Robust Teachers

2026-02-02 · Alessandro De Palma arxiv

Adversarial training attains strong empirical robustness to specific adversarial attacks by training on concrete adversarial perturbations, but it produces neural networks that are not amenable to strong robustness certi…

Knowledge Distillation

PointCert: Point Cloud Classification with Deterministic Certified Robustness Guarantees

2023-03-03 · CVPR 2023 1 · Jinghuai Zhang, Jinyuan Jia, Hongbin Liu, Neil Zhenqiang Gong

Point cloud classification is an essential component in many security-critical applications such as autonomous driving and augmented reality. However, point cloud classifiers are vulnerable to adversarially perturbed poi…

Autonomous DrivingClassificationPoint Cloud Classification

Adversarially Learned Anomaly Detection

2018-12-06 · Houssam Zenati, Manon Romain, Chuan Sheng Foo, Bruno Lecouat 외

Anomaly detection is a significant and hence well-studied problem. However, developing effective anomaly detection methods for complex and high-dimensional data remains a challenge. As Generative Adversarial Networks (GA…

Anomaly Detection

PatchCleanser: Certifiably Robust Defense against Adversarial Patches for Any Image Classifier

2021-08-20 · Chong Xiang, Saeed Mahloujifar, Prateek Mittal

The adversarial patch attack against image classification models aims to inject adversarially crafted pixels within a restricted image region (i.e., a patch) for inducing model misclassification. This attack can be reali…

image-classificationImage Classification