Speaker：Xin Li, Associate Vice-Chancellor for Graduate Studies and Research, Duke Kunshan University
Dean of Graduate Studies, Duke Kunshan University
Professor of Electrical and Computer Engineering, Duke University
Professor of Electrical and Computer Engineering.Duke Kunshan University
Inviter：Mian Li
Time：14:30 - 15:30, November 3, 2023 (Beijing Time)
Location：Room 200, Yue-Kong Pao Library's Annex
Abstract：
Wafer classification is a critical task for semiconductor manufacturing. Most conventional algorithms require a large-scale perfectly-labeled dataset to train accurate classifiers. In practice, it is usually difficult or even impossible to collect perfect labels without errors, and the classification accuracy in the presence of imperfectly labeled data may substantially degrade. In this presentation, we will discuss a number of novel techniques to facilitate robust wafer classification with noisy labels. These techniques can be classified into three broad categories: (1) data cleaning methods, (2) loss-function-based methods, and (3) co-teaching methods. The efficacy of robust wafer classification will be demonstrated by several industrial datasets.
Biography：
Xin Li received the Ph.D. degree in Electrical & Computer Engineering from Carnegie Mellon University in 2005. He is currently a Professor in ECE at Duke University and serves as the Associate Vice-Chancellor for Graduate Studies and Research at Duke Kunshan University. His research interests include integrated circuits, signal processing and data analytics. Dr. Li was the Deputy Editor-in-Chief of IEEE TCAD. He was an Associate Editor of IEEE TCAD, IEEE TBME, ACM TODAES, IEEE D&T and IET CPS. He was the General Chair of ISVLSI and FAC. He received the NSF CAREER Award in 2012 and six Best Paper Awards from IEEE TCAD, DAC, ICCAD and ISIC. He is a Fellow of IEEE.
