主 题: A Statistical Journey through Trustworthy AI
主讲人:加州大学洛杉矶分校程光教授
主持人:统计学院林华珍教授
时间:2022年5月11日(周三)上午9:00-10:00
直播平台及会议ID:腾讯会议,ID: 948-189-942
主办单位:统计研究中心和统计学院 科研处
主讲人简介:
程光,加州大学洛杉矶分校的统计学教授。他于2006年在威斯康星大学麦迪逊分校获得统计学博士学位。主要研究方向为可信AI、深度学习理论和统计机器学习。他是IMS的 fellow,并获得以下奖项:NSF CAREER、Noether Young Scholar和Simons Fellowship in Mathematics。现为JASA -- T&M, Canadian Journal of Statistics, Journal of Blockchain Research的副主编。
更多详情请见其个人主页:http://www.stat.ucla.edu/~guangcheng/
内容提要:
Our lab believes that the next generation of AI is mainly driven by trustworthiness, beyond performance. This talk attempts to offer statistical solutions to embrace three challenges in trustworthy AI: privacy, robustness and fairness. Specifcally, we consider privacy protection by machine un-learning, enhanced adversarial robustness by utilizing unlabelled data, and establishing fair Bayes-optimal classifiers. These results demonstrate the unique value of statisticians in studying trustworthy AI from empirical, methodological or theoretical aspects.