The Community for Technology Leaders
2018 IEEE Symposium on Security and Privacy (SP) (2018)
San Francisco, CA, US
May 21, 2018 to May 23, 2018
ISSN: 2375-1207
ISBN: 978-1-5386-4353-2
pp: 948-963
Timon Gehr , ETH Zurich
Matthew Mirman , ETH Zurich
Petar Tsankov , ETH Zurich
Swarat Chaudhuri , Rice University
Martin Vechev , ETH Zurich
We present AI2, the first sound and scalable analyzer for deep neural networks. Based on overapproximation, AI2 can automatically prove safety properties (e.g., robustness) of realistic neural networks (e.g., convolutional neural networks). The key insight behind AI2 is to phrase reasoning about safety and robustness of neural networks in terms of classic abstract interpretation, enabling us to leverage decades of advances in that area. Concretely, we introduce abstract transformers that capture the behavior of fully connected and convolutional neural network layers with rectified linear unit activations (ReLU), as well as max pooling layers. This allows us to handle real-world neural networks, which are often built out of those types of layers. We present a complete implementation of AI2 together with an extensive evaluation on $20$ neural networks. Our results demonstrate that: (i) AI2 is precise enough to prove useful specifications (e.g., robustness), (ii) AI2 can be used to certify the effectiveness of state-of-the-art defenses for neural networks, (iii) AI2 is significantly faster than existing analyzers based on symbolic analysis, which often take hours to verify simple fully connected networks, and (iv) AI2 can handle deep convolutional networks, which are beyond the reach of existing methods.

T. Gehr, M. Mirman, D. Drachsler-Cohen, P. Tsankov, S. Chaudhuri and M. Vechev, "AI2: Safety and Robustness Certification of Neural Networks with Abstract Interpretation," 2018 IEEE Symposium on Security and Privacy (SP), San Francisco, CA, US, , pp. 948-963.
79 ms
(Ver 3.3 (11022016))