讲座题目：Harmonizing Fairness with Utility in Data and Learning
时间：2023年5月30日上午10点
地点：新主楼A1128
主讲人：HongfuLiu
主讲人信息：
Dr. Hongfu Liu is an Assistant Professor of Computer Science at Brandeis University. His research interests lie in core machine learning and AI-assisted applications. He has published over 100 papers (e.g., KDD, NeurIPS, ICLR, ICML, IJCAI, AAAI, ICDM, SDM, CIKM, CVPR, ICCV, TPAMI, and TKDE). These publications have received over 2,800 citations with an h-index of 30 according to Google Scholar as of May 2023. He has also won several awards including the First Place Award in MS-Celel-1M Grand Challenge in ICCV 2017, the NVIDIA CCS Best Student Paper Award in FG 2021, the 2021 INNS Aharon Katzir Young Investigator Award, the top reviewer in UAI 2022, the highlighted Area Chair in ICLR 2022, the 2022 Global Top-25 Chinese Young Scholars in AI (Data Mining Area) by Baidu Scholar and the notable Area Chair in ICLR 2023. He has served as an Associate Editor of IEEE CIM and as an Area Chair of ICLR, ICML, and NeurIPS.
讲座主要内容：
Utility-oriented machine learning systems have made their way into the real world, assisting with high-stakes decision-making in areas such as college admissions, loan approvals, and credit auditing. However, the inherent biases embedded in these models, stemming from unbalanced data, can backfire and worsen existing social barriers. Fairness in machine learning has gained significant attention to tackle the above algorithmic discrimination in diverse tasks. Among this literature, mitigating unfairness can be achieved through three main categories of solutions: pre-processing, in-processing, and post-processing. In this talk, I will introduce our recent studies on fairness learning in terms of learning and data perspectives.
