PROOFS 2021

The workshop takes place on Friday September 17, in the European afternoon, after the end of CHES. All times in the schedule are UTC times. You can access the papers by clicking on the titles.

Abstract: Unlike traditional programs, AI systems built on neural networks are subject to a variety of vulnerabilities, such as adversarial attacks and backdoor attacks. While it is tempting to apply existing program analysis techniques to address the AI problems, it should be also noted that neural networks are fundamentally different from software systems, and thus what worked for software may not work for neural networks. In this talk, I will discuss what perhaps would be required to build trustworthy AI. Our effort on establishing the absence of backdoor attacks on a given neural network will be presented as a concrete example.

Sun Jun is currently an associate professor at Singapore Management University (SMU). He received Bachelor and PhD degrees in computing science from National University of Singapore (NUS) in 2002 and 2006. In 2007, he received the prestigious LEE KUAN YEW postdoctoral fellowship. He has been a faculty member since 2010 and a visiting scholar at MIT from 2011-2012. Jun's research interests include software engineering, formal methods, program analysis and cyber-security. He is the co-founder of the PAT model checker. To this date, he has published more than 250 journal articles or peer-reviewed conference papers, and won many best paper awards.