Please note: This seminar will be given online.
Hongyang Zhang, Postdoctoral Fellow
Toyota Technological Institute at Chicago
Deep learning models are often vulnerable to adversarial examples. In this talk, we will focus on robustness and security of machine learning against adversarial examples. There are two types of defenses against such attacks: 1) empirical and 2) certified adversarial robustness.