Recent breakthroughs in Computer Vision, Natural Language Processing, and Speech Recognition are bringing Machine Learning (ML) models into the center of security-critical applications including autonomous driving, face recognition, and healthcare. These developments make security aspects of ML increasingly important.
Join MIT PhD candidate Hadi Salman in a FREE talk about “Adversarial Attacks and Defenses in Machine Learning” next Monday, October 24th from 7:00 to 8:00 PM (GMT+3).
During this free talk, Hadi will:
Give an overview of one serious security vulnerability in ML models, namely their susceptibility to adversarial examples.
Cover various methods to generate such adversarial examples, aka adversarial attacks.
Go over strategies to defend against these attacks.
Answer your questions about the topic.
P.S. The talk requires no prior knowledge beyond basic understanding of Machine Learning.