SIAM News Blog

UQ22 Prize Spotlight: Mengyang Gu

Mengyang Gu is the 2022 recipient of the SIAM Activity Group on Uncertainty Quantification Early Career Prize. The award will be presented at the 2022 SIAM Conference on Uncertainty Quantification (UQ22), to be held in a hybrid format April 12 – 15, 2022. The in-person component of the conference will be held in Atlanta, Georgia, United States. Dr. Gu will present a talk at the conference titled “Scalable Gaussian Process for Computer Model Emulation and Uncertainty Quantification” on Wednesday, April 13, 2022 at 10:40 a.m. ET.

The prize is awarded to Dr. Gu for his contributions to the analysis and estimation of Gaussian process emulators.

The SIAM Activity Group on Uncertainty Quantification awards this prize every two years to one individual in their early career for outstanding research contributions in the field of uncertainty quantification in the three calendar years prior to the award year.

Dr. Gu is an assistant professor in the Department of Statistics and Applied Probability at UC Santa Barbara. He obtained his Ph.D. and M.S. degrees in Statistical Science from Duke University, advised by Professor James O. Berger. Prior to joining UCSB, he was a research assistant professor in the Department of Applied Mathematics and Statistics at Johns Hopkins University. His research interests include Bayesian analysis, computer model emulation, inverse problem, spatio-temporal modeling, and image analysis with applications in geological hazard quantification, material design, and characterization. 

Q: Why are you excited to receive the SIAG/UQ Early Career Prize?

A: I am exceedingly delighted to receive this prize and to be recognized by the broad community for my research in uncertainty quantification. I am grateful to my advisor and collaborators for their insightful suggestions and inspiration. This award greatly motivates me to solve challenging and exciting research problems. 

Q: Could you tell us a bit about the research that won you the prize? 

A: The main contribution of my work is on developing efficient and robust emulators to approximate computationally expensive computer simulations. Computer simulations are widely used for advancing scientific discovery when actual experiments are costly or impossible. For instance, mechanical models are frequently used to model earth systems for geological hazard assessment, while molecular dynamic simulation is one of the most useful tools for predicting a material’s property. The computational complexity of these models increases dramatically with the dimension of parameter space, making these simulations prohibitively slow. To address this problem, I developed fast Gaussian process emulators to approximate physics simulations at massive coordinates, expanding the complexity model space researchers can simulate. The particular importance of my work is on scalable computation of statistical inference from massive amounts of data, and robust prediction with uncertainty quantification.

Q: What does your work mean to the public?

A: Data science techniques and algorithms facilitate scientific discovery and guide policy decisions. For instance, a few years ago, I collaborated with scientists in the U.S. Geological Survey for monitoring geological hazards, such as the 2018 eruption of Kīlauea on the island of Hawai’i, by ground deformation data including satellite interferograms and GPS observations. These findings can help guide policy decisions and save lives. Computational algorithms, such as the Kalman filter and fast Fourier transform, have become standard tools ubiquitously used in research and in daily life. My goal is to develop data science algorithms as efficient and robust as these tools for everyone to use.

Q: What does being a member of SIAM mean to you?

A: SIAM is a wonderful community. I attended the first SIAM Conference on Uncertainty Quantification in 2014, and then I attended all the following SIAM UQ conferences. I am grateful for SIAM's efforts in promoting the development of applied mathematical and computational tools in solving real-world problems through interdisciplinary research. I am honored to be a part of this effort. I will continue to be an active participant in events organized by SIAM, contributing on research products, and educating the next generation of applied mathematicians and data scientists.

Learn more about the SIAM Conference on Uncertainty Quantification. Registration for in-person participation is open through April 5, 2022, and virtual participation is open through the end of the conference on April 15, 2022.

blog comments powered by Disqus