Adversarial example or just noise?
For untargeted attacks, how does one know if the decrease in accuracy is due to an adversarial perturbation or its just being attributed to adding noise to input.
Topic adversarial-ml cnn
Category Data Science
About
Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.