Deep Neural Network Fingerprinting by Conferrable Adversarial Examples

TL;DR

This paper introduces a novel fingerprinting technique using conferrable adversarial examples to reliably identify surrogate models in machine learning as a service, demonstrating high robustness against various attacks.

Contribution

It proposes a new method to generate conferrable adversarial examples for model fingerprinting and shows its effectiveness and robustness against multiple attack strategies.

Findings

Achieves ROC AUC of 1.0 in surrogate verification

Robust against fine-tuning, pruning, retraining, and transfer learning

First method to reach perfect verification accuracy

Abstract

In Machine Learning as a Service, a provider trains a deep neural network and gives many users access. The hosted (source) model is susceptible to model stealing attacks, where an adversary derives a surrogate model from API access to the source model. For post hoc detection of such attacks, the provider needs a robust method to determine whether a suspect model is a surrogate of their model. We propose a fingerprinting method for deep neural network classifiers that extracts a set of inputs from the source model so that only surrogates agree with the source model on the classification of such inputs. These inputs are a subclass of transferable adversarial examples which we call conferrable adversarial examples that exclusively transfer with a target label from a source model to its surrogates. We propose a new method to generate these conferrable adversarial examples. We present an extensive study on the irremovability of our fingerprint against fine-tuning, weight pruning, retraining, retraining with different architectures, three model extraction attacks from related work, transfer learning, adversarial training, and two new adaptive attacks. Our fingerprint is robust against distillation, related model extraction attacks, and even transfer learning when the attacker has no access to the model provider's dataset. Our fingerprint is the first method that reaches a ROC AUC of 1.0 in verifying surrogates, compared to a ROC AUC of 0.63 by previous fingerprints.