Naresh R. Shanbhag

Generalized Depthwise-Separable Convolutions for Adversarially Robust and Efficient Neural Networks

2021 · arXiv (Cornell University)

Despite their tremendous successes, convolutional neural networks (CNNs) incur high computational/storage costs and are vulnerable to adversarial perturbations. Recent works on robust model compression address these challenges by combining model compression techniques with adversarial training. …