Report Title: The Road to AI Fairness

Report time: April 29th, 10:00-11:00 Offline report location: B1002, Science Building

Host: Professor Pugeguang

Report Summary:

Neural networks have been widely used in many systems, including those related to social resource allocation, such as credit scoring and resume selection. Worryingly, scientists have discovered many fairness issues in neural networks. Compared to traditional software, these problems on neural networks are less easily discovered or solved because people have no way to understand how neural networks work. So how can we ensure that neural networks that are crucial to fairness are trustworthy? In this lecture, I will discuss some issues related to fairness in neural networks, the latest research, and solutions.

Reported by:

Sun Jun. site is currently a lifelong professor at the Singapore Management University (SMU). He obtained a Bachelor's and Doctoral degree in Computer Science from the National University of Singapore in 2002 and 2006. In 2007, he was awarded the Lee Kuan Yew Postdoctoral Scholarship. He has been employed as a professor since 2010. He served as a visiting scholar at the Massachusetts Institute of Technology from 2011 to 2012. Sun Jun's research interests include network security, software engineering, and formal methods. He has published over 250 articles and conference papers, including top conferences in multiple fields. He has released multiple software analysis tools, including the PAT model checker used by multiple companies. He is also a senior technical consultant for multiple companies.