Accelerated Smoothing: A Scalable Approach to Randomized Smoothing
This addresses scalability issues for practitioners using certifiable defenses against adversarial attacks, though it is an incremental improvement on existing methods.
The paper tackles the computational bottleneck in randomized smoothing for adversarial defense by replacing Monte Carlo sampling with a surrogate neural network, achieving a nearly 600X speedup in robust radius certification.
Randomized smoothing has emerged as a potent certifiable defense against adversarial attacks by employing smoothing noises from specific distributions to ensure the robustness of a smoothed classifier. However, the utilization of Monte Carlo sampling in this process introduces a compute-intensive element, which constrains the practicality of randomized smoothing on a larger scale. To address this limitation, we propose a novel approach that replaces Monte Carlo sampling with the training of a surrogate neural network. Through extensive experimentation in various settings, we demonstrate the efficacy of our approach in approximating the smoothed classifier with remarkable precision. Furthermore, we demonstrate that our approach significantly accelerates the robust radius certification process, providing nearly $600$X improvement in computation time, overcoming the computational bottlenecks associated with traditional randomized smoothing.