Jialei Chen Wins Two Best Student Paper Awards at the 2020 INFORMS Conference

<p><strong>Jialei Chen</strong>, a Ph.D. student in the H. Stewart School of Industrial &amp; Systems Engineering at Georgia Tech.</p>

Jialei Chen, a Ph.D. student in the H. Stewart School of Industrial & Systems Engineering at Georgia Tech.

Jialei Chen, a doctoral student in the H. Milton Stewart School of Industrial and Systems Engineering and a graduate research assistant in the Georgia Tech Manufacturing Institute (GTMI), won two Best Student Paper Awards at this year’s 2020 INFORMS Conference. The annual INFORMS conference on business analytics and operations research brings together nearly 1,000 leading analytics professionals and industry experts to share ideas, network and learn about a range of current topics and trends that can help businesses and organizations improve their analytics prowess by applying science to the art of business.

Chen won the Best Student Paper Award in the Quality, Statistics, and Reliability track for “Adaptive Design for Gaussian Process Regression under Censoring.” This paper presented an experimental design and modeling method for censored physical experiments. Censoring is commonly encountered in experimentation due to the limits in a measurement device, safety considerations of the experimenter, and a fixed experimental time budget. To tackle this, he proposed a novel adaptive design method, which first estimates the possibility of censoring and then adaptively chooses design points to minimize predictive uncertainty under censoring. He demonstrates the effectiveness of the proposed method in two real-world applications on surgical planning and wafer manufacturing.

Chen received the Best Student Paper Runner-up Award in the Data Mining track for “APIK: A Physics-Informed Kriging Model with Partial Differential Equations.” This paper presented a learning framework that combines limited data and the auxiliary partial differential equations. One of the key challenges in applying state-of-the-art machine learning methods in real-world engineering applications is that the available measurement data is scarce. In this work, he proposed to incorporate the auxiliary partial differential equations in the learning model and therefore improve the predive performance. The proposed APIK model can leverage linear and nonlinear PDEs and enjoy simple and closed-form prediction and uncertainty quantification. He applied the proposed method to two real-world applications on flow dynamics and thermal processes.

Chen’s advisors for both papers are A. Russell Chandler III Professor Roshan Joseph and Harold E. Smalley Professor Chuck Zhang.

“I’m honored to have won best student paper award in the quality, statistics, and reliability track at INFORMS 2020, and to have another paper win second place in the data mining track,” said Chen. “My research focuses on engineering-driven learning methodologies, and data-driven modeling for complex engineering and manufacturing systems. The two awards are a great encouragement for me and inspire me to accomplish more in-depth and impactful works in the future. I would like to express my highest gratitude to my supervisors, professors Chuck Zhang and Roshan Joseph. I would also like to thank the support and assistance from GTMI, which helped to make the two projects possible.”

News Contact

Walter Rich