SMMG Seminar Series Engineering-informed Machine Learning for Smart Manufacturing - Engineering-informed Transfer Learning for Model Transfer in Smart Manufacturing – I
Supporting the below United Nations Sustainable Development Goals:支持以下聯合國可持續發展目標：支持以下联合国可持续发展目标：
Abstract:
A fundamental issue in engineering is whether a model developed in a study under one setting can be successfully transferred and applied to new settings. Bottlenecks to model transfer are introduced by lurking variables, or factors whose levels are completely unobserved due either to negligence, infeasibility of measurement, or insufficient knowledge. This talk presents a scheme of model transfer based on engineering effect equivalence concept. The total equivalent amount of the lurking variables is represented in terms of an observed base factor with respect to a model feature. The first part of this talk will demonstrate the effect-equivalence based transfer learning and control in traditional manufacturing.
Dr. Qiang S. Huang is a Professor at the Daniel J. Epstein Department of Industrial and Systems Engineering, University of Southern California (USC), Los Angeles. He is currently a visiting Professor at SMMG. His research focuses on Machine Learning for Additive Manufacturing (ML4AM) and quality control theory and methods for personalized manufacturing. He was the holder of the Gordon S. Marshall Early Career Chair in Engineering at USC from 2012 to 2016. He received IISE Fellow Award, ASME Fellow, NSF CAREER award, the 2021 IEEE CASE Best Conference Paper Award, 2013 IEEE Transactions on Automation Science and Engineering Best Paper Award, among others. He has served as a Department Editor for IISE Transactions and an Associate Editor for ASME Transactions, Journal of Manufacturing Science and Engineering.