- Abstract: Robustness of neural networks has recently been highlighted by the adversarial examples, i.e., inputs added with well-designed perturbations which are imperceptible to humans but can cause the network to give incorrect outputs. In this paper, we design a new CNN architecture that by itself has good robustness. We introduce a simple but powerful technique, Random Mask, to modify existing CNN structures. We show that CNN with Random Mask achieves state-of-the-art performance against black-box adversarial attacks without applying any adversarial training. We next investigate the adversarial examples which “fool” a CNN with Random Mask. Surprisingly, we find that these adversarial examples often “fool” humans as well. This raises fundamental questions on how to define adversarial examples and robustness properly.
- Keywords: adversarial examples, robust machine learning, cnn structure, metric, deep feature representations
- TL;DR: We propose a technique that modifies CNN structures to enhance robustness while keeping high test accuracy, and raise doubt on whether current definition of adversarial examples is appropriate by generating adversarial examples able to fool humans.