Fast and Effective Robustness Certification

Gagandeep Singh
Gagandeep Singh
Timon Gehr
Timon Gehr
Matthew Mirman
Matthew Mirman

ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 31 (NIPS 2018), pp. 10825-10836, 2018.

Cited by: 95|Views9
EI

Abstract:

We present a new method and system, called DeepZ, for certifying neural network robustness based on abstract interpretation. Compared to state-of-the-art automated verifiers for neural networks, DeepZ: (i) handles ReLU, Tanh and Sigmoid activation functions, (ii) supports feedforward, convolutional, and residual architectures, (iii) is si...More

Code:

Data:

Your rating :
0

 

Tags
Comments