Adversarial Robustness of Tiny-ML models

Tiny ML has a security problem - An Adversarial Perturbation Prospective