Fastened CROWN: Tightened Neural Network Robustness Certificates
This work addresses safety concerns in deep learning by improving robustness certification for neural networks, though it appears incremental as it builds on existing CROWN methods.
The paper tackles the problem of certifying neural network robustness by demonstrating the optimality of deterministic CROWN solutions under mild constraints, eliminating the need for computationally expensive linear programming. It proposes FROWN, an optimization-based algorithm that tightens robustness certificates, with experiments showing effectiveness in safeguarding larger robust regions.
The rapid growth of deep learning applications in real life is accompanied by severe safety concerns. To mitigate this uneasy phenomenon, much research has been done providing reliable evaluations of the fragility level in different deep neural networks. Apart from devising adversarial attacks, quantifiers that certify safeguarded regions have also been designed in the past five years. The summarizing work of Salman et al. unifies a family of existing verifiers under a convex relaxation framework. We draw inspiration from such work and further demonstrate the optimality of deterministic CROWN (Zhang et al. 2018) solutions in a given linear programming problem under mild constraints. Given this theoretical result, the computationally expensive linear programming based method is shown to be unnecessary. We then propose an optimization-based approach \textit{FROWN} (\textbf{F}astened C\textbf{ROWN}): a general algorithm to tighten robustness certificates for neural networks. Extensive experiments on various networks trained individually verify the effectiveness of FROWN in safeguarding larger robust regions.