Cybersecurity

The goal of the course is to teach students how to apply machine learning (including deep learning) techniques in cybersecurity, and understand their vulnerabilities in adversarial settings. Specific Objectives The students will learn formally and practically how machine learning models work, their applications to cybersecurity problems, their vulnerabilities, existing attacks and mitigation techniques. Knowledge and Understanding - knowledge and understanding of the mathematical foundations behind modern machine-learning techniques - knowledge and understanding of the vulnerability of modern machine-learning techniques to adversarial attacks - knowledge and understanding of state-of-the-art mitigation techniques against these attacks - knowledge and understanding of various applications of machine learning to cybersecurity problems Autonomy of Judgement The students will be able to assess the security of machine-learning applications and to evaluate possible failure modes and vulnerabilities to adversarial attacks Students will be able to describe the security and appropriateness of machine learning applications, and appropriately present and discuss potential failure modes Next Study Abilities Students will be prepared to understand more complex machine-learning models and techniques, and will be equipped with the necessary knowledge to pursue open research problems in the areas of machine learning and cybersecurity