Detecting Adversarial Examples via Neural Fingerprinting

Abstract

Deep neural networks are vulnerable to adversarial examples, which dramatically alter model output using small input changes. We propose Neural Fingerprinting, a simple, yet effective method to detect adversarial examples by verifying whether model behavior is consistent with a set of secret fingerprints, inspired by the use of biometric and cryptographic signatures. The benefits of our method are that 1) it is fast, 2) it is prohibitively expensive for an attacker to reverse-engineer which fingerprints were used, and 3) it does not assume knowledge of the adversary. In this work, we pose a formal framework to analyze fingerprints under various threat models, and characterize Neural Fingerprinting for linear models. For complex neural networks, we empirically demonstrate that Neural Fingerprinting significantly improves on state-of-the-art detection mechanisms by detecting the strongest known adversarial attacks with 98-100% AUC-ROC scores on the MNIST, CIFAR-10 and MiniImagenet (20 classes) datasets. In particular, the detection accuracy of Neural Fingerprinting generalizes well to unseen test-data under various black- and whitebox threat models, and is robust over a wide range of hyperparameters and choices of fingerprints.

First 2 authors contributed equally

References in corpus (10)

• Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samples
• A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks
• Certified Defenses against Adversarial Examples
• Adversarially Robust Generalization Requires More Data
• Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionality
• Achieving Human Parity in Conversational Speech Recognition
• Adversarial Risk and the Dangers of Evaluating Against Weak Attacks
• Detecting Adversarial Samples from Artifacts
• MagNet and "Efficient Defenses Against Adversarial Attacks" are Not Robust to Adversarial Examples
• Defense against Adversarial Attacks Using High-Level Representation Guided Denoiser

Cited by in corpus (4)

• Towards resilient machine learning for ransomware detection
• Adversarial Defense by Latent Style Transformations
• TEAM: We Need More Powerful Adversarial Examples for DNNs
• MadNet: Using a MAD Optimization for Defending Against Adversarial Attacks