Found 1 document(s)
Page 1 of 1
-
Deep neural networks (NNs) for computer vision are vulnerable to adversarial attacks, such that minuscule malicious changes to inputs may induce unintuitive outputs. One key approach to verify and mitigate such robustnes...