About this course

Our society is relying extensively on artificial intelligence, in many cases, using neural networks. Artificial intelligence comes with great opportunities, but also certain safety challenges. For example, it is important that those neural networks are robust against so-called adversarial attacks. In this course, Dr. Jan N. van Rijn (Leiden University) will explain various concepts about safety in AI that are important to consider when deploying a neural network in practice. In particular, you will learn what kind of methods can be used to verify whether a network is robust against adversarial attacks and how Automated Machine Learning can be used to further improve these methods.