Enhancing Certified Robustness via Smoothed Weighted Ensembling
This work addresses a key bottleneck in adversarial robustness for machine learning systems, offering an incremental improvement over existing randomized smoothing methods.
The paper tackles the problem of finding optimal base classifiers for randomized smoothing to enhance certified robustness against adversarial attacks, showing that their Smoothed Weighted Ensembling (SWEEN) scheme outperforms candidate models by a large margin and achieves comparable performance to large models with reduced training time.
Randomized smoothing has achieved state-of-the-art certified robustness against $l_2$-norm adversarial attacks. However, it is not wholly resolved on how to find the optimal base classifier for randomized smoothing. In this work, we employ a Smoothed WEighted ENsembling (SWEEN) scheme to improve the performance of randomized smoothed classifiers. We show the ensembling generality that SWEEN can help achieve optimal certified robustness. Furthermore, theoretical analysis proves that the optimal SWEEN model can be obtained from training under mild assumptions. We also develop an adaptive prediction algorithm to reduce the prediction and certification cost of SWEEN models. Extensive experiments show that SWEEN models outperform the upper envelope of their corresponding candidate models by a large margin. Moreover, SWEEN models constructed using a few small models can achieve comparable performance to a single large model with a notable reduction in training time.