Randomized smoothing-based certification is an effective approach for obtaining robustness certificates of deep neural networks (DNNs) against adversarial attacks. This method constructs a smoothed DNN model and certifies its robustness through statistical sampling, but it is computationally expensive, especially when certifying with a large number of samples. Furthermore, when the smoothed model is modified (e.g., quantized or pruned), certification guarantees may not hold for the modified DNN, and recertifying from scratch can be prohibitively expensive. We present the first approach for incremental robustness certification for randomized smoothing, IRS. We show how to reuse the certification guarantees for the original smoothed model to certify an approximated model with very few samples. IRS significantly reduces the computational cost of certifying modified DNNs while maintaining strong robustness guarantees. We experimentally demonstrate the effectiveness of our approach, showing up to 3x certification speedup over the certification that applies randomized smoothing of the approximate model from scratch.

翻译：基于随机平滑的认证是一种有效的方法，用于获取深度神经网络（DNN）对抗攻击的鲁棒性证书。该方法通过统计采样构建平滑后的DNN模型并认证其鲁棒性，但计算成本高昂，尤其是在使用大量样本进行认证时。此外，当平滑后的模型被修改（例如量化或剪枝）时，认证保证可能不再适用于修改后的DNN，而从头重新认证的成本难以承受。我们提出了首个针对随机平滑的增量式鲁棒性认证方法IRS。我们展示了如何复用原始平滑模型的认证保证，以极少的样本对近似模型进行认证。IRS在保持强鲁棒性保证的同时，显著降低了认证修改后DNN的计算成本。实验结果表明，我们的方法有效，与直接对近似模型从头应用随机平滑的认证相比，认证速度可提升3倍。