Training Ensembles to Detect Adversarial Examples

2017/12/11 by Alexander Bagnall, Răzvan Bunescu, Bagnall, Alexander +3 · 4 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · #Advanced Malware Detection Techniques #Adversarial Robustness in Machine Learning #Bacillus and Francisella bacterial research #Computer Vision and Pattern Recognition (cs.CV) #Cryptography and Security (cs.CR) #FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · pdf · doi:10.48550/arxiv.1712.04006

openalex publication_date 2017/12/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We propose a new ensemble method for detecting and classifying adversarial examples generated by state-of-the-art attacks, including DeepFool and C&W. Our method works by training the members of an ensemble to have low classification error on random benign examples while simultaneously minimizing agreement on examples outside the training distribution. We evaluate on both MNIST and CIFAR-10, against oblivious and both white- and black-box adversaries.

Cited by

• Evaluating Ensemble Robustness Against Adversarial Attacks
• SHIELD: Secure Hypernetworks for Incremental Expansion Learning Defense
• PAC-learning in the presence of evasion adversaries
• Improving Adversarial Robustness of Ensembles with Diversity Training
• Exploiting the Inherent Limitation of L0 Adversarial Examples

Related

• GAT: Generative Adversarial Training for Adversarial Example Detection and Robust Classification
• Towards Robust Detection of Adversarial Examples
• Ensemble Adversarial Training: Attacks and Defenses
• Beating Attackers At Their Own Games: Adversarial Example Detection Using Adversarial Gradient Directions
• Guardian of the Ensembles: Introducing Pairwise Adversarially Robust Loss for Resisting Adversarial Attacks in DNN Ensembles