Evaluating Robustness of Neural Networks

The robustness of neural networks to adversarial examples has received great attention due to security implications. Despite various attack approaches to crafting visually imperceptible adversarial examples, little has been developed towards a comprehensive measure of robustness. In this talk, I'll present a series of our work on robustness evaluation and certification, including the first robustness score CLEVER, efficient certification algorithms Fast-Lin, CROWN, CNN-Cert, and probabilistic robustness verification algorithm PROVEN. Our proposed approaches are computationally efficient and provide good quality of robustness estimate/certificate as demonstrated by extensive experiments on MNIST, CIFAR and ImageNet.