PRADA: Protecting against DNN Model Stealing Attacks

Abstract

Machine learning (ML) applications are increasingly prevalent. Protecting the confidentiality of ML models becomes paramount for two reasons: (a) a model can be a business advantage to its owner, and (b) an adversary may use a stolen model to find transferable adversarial examples that can evade classification by the original model. Access to the model can be restricted to be only via well-defined prediction APIs. Nevertheless, prediction APIs still provide enough information to allow an adversary to mount model extraction attacks by sending repeated queries via the prediction API. In this paper, we describe new model extraction attacks using novel approaches for generating synthetic queries, and optimizing training hyperparameters. Our attacks outperform state-of-the-art model extraction in terms of transferability of both targeted and non-targeted adversarial examples (up to +29-44 percentage points, pp), and prediction accuracy (up to +46 pp) on two datasets. We provide take-aways on how to perform effective model extraction attacks. We then propose PRADA, the first step towards generic and effective detection of DNN model extraction attacks. It analyzes the distribution of consecutive API queries and raises an alarm when this distribution deviates from benign behavior. We show that PRADA can detect all prior model extraction attacks with no false positives.

arXiv comment17 pages, 7 figures, 9 tables. Accepted for publication in the 4th IEEE European Symposium on Security and Privacy (EuroS&P 2019)

References in corpus (8)

• Distilling the Knowledge in a Neural Network
• Practical Bayesian Optimization of Machine Learning Algorithms
• Federated Learning: Strategies for Improving Communication Efficiency
• Ensemble Adversarial Training: Attacks and Defenses
• ZOO: Zeroth Order Optimization based Black-box Attacks to Deep Neural Networks without Training Substitute Models
• Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examples
• Detecting Adversarial Samples from Artifacts
• On the (Statistical) Detection of Adversarial Examples

Cited by in corpus (5)

• Stealing Neural Networks via Timing Side Channels
• Fusion: Efficient and Secure Inference Resilient to Malicious Servers
• IPGuard: Protecting Intellectual Property of Deep Neural Networks via Fingerprinting the Classification Boundary
• Towards Security Threats of Deep Learning Systems: A Survey
• ArchNet: Data Hiding Model in Distributed Machine Learning System