Can collaborative learning be private, robust and scalable?. (arXiv:2205.02652v1 [cs.LG])

Web: http://arxiv.org/abs/2205.02652

We investigate the effectiveness of combining differential privacy, model
compression and adversarial training to improve the robustness of models
against adversarial samples in train- and inference-time attacks. We explore
the applications of these techniques as well as their combinations to determine
which method performs best, without a significant utility trade-off. Our
investigation provides a practical overview of various methods that allow one
to achieve a competitive model performance, a significant reduction in model's
size and an improved empirical adversarial robustness …

arxiv collaborative learning scalable