The (In)security of Today’s Machine Learning Systems
Today's technology companies are rushing to adopt and deploy systems that make use of machine learning, and in particular deep learning (artificial neural networks). Prototypes are in development for everything from facial recognition security systems for buildings to traffic sign recognition systems for autonomous vehicles. But is the technology ready? In this lecture, Ben Zhao will present recent work from his lab on fundamental weaknesses in deep learning systems that make them easy to compromise, as well as some proposed defenses.
Can’t be there in person? Check out the livestream on Facebook.
I am Neubauer Professor of Computer Science at University of Chicago. My research covers a range of topics from large-distributed networks and systems, HCI, security and privacy, and wireless / mobile systems, mostly from a data-driven perspective. My current projects are focused on three areas: data-driven models of user behavior/interactions, security of online and mobile communities, and wireless systems and protocols. My work targets a range of top conferences, including WWW/IMC, UsenixSecurity/NDSS/S&P/CCS, CHI/CSCW, and Mobicom/SIGCOMM/NSDI.