Abstract
We investigate the relationship between the frequency spectrum of image data
and the generalization behavior of convolutional neural networks (CNN). We
first notice CNN's ability in capturing the high-frequency components of
images. These high-frequency components are almost imperceptible to a human.
Thus the observation leads to multiple hypotheses that are related to the
generalization behaviors of CNN, including a potential explanation for
adversarial examples, a discussion of CNN's trade-off between robustness and
accuracy, and some evidence in understanding training heuristics. Our
observation also immediately inspire methods related to the adversarial attack
and defense methods.
Users
Please
log in to take part in the discussion (add own reviews or comments).