Elizabeth Qian
Assistant Professor at Georgia Tech
I'm an Assistant Professor at Georgia Tech in the Schools of Aerospace Engineering and Computational Science and Engineering. My research develops mathematical and computational methods that enable engineers to make better design decisions faster. My specialties are model reduction, data-driven modeling, scientific machine learning, and multi-fidelity methods. You can learn more on my research page.
Prior to joining the faculty at Georgia Tech, I held a von Kármán Instructorship at Caltech in the Department of Computing + Mathematical Sciences. I received my SB, SM, and PhD degrees from the MIT Department of Aeronautics & Astronautics. I also currently hold a visiting appointment as a Hans Fischer Fellow at the Technical University of Munich.
I am excited about mentoring and teaching the next generation of aerospace engineers and computational scientists, and I work to make my professional communities more equitable, diverse, and inclusive for generations to come. My service and teaching contributions have previously been recognized with departmental and division-wide DEI awards, as well as an institute-wide teaching award.
Upcoming talks & activities
October 2024: I will present our recent work on multifidelity machine learning from scarce data on October 10 in the 3pm ET webinar organized by the USACM technical thrust area on Uncertainty Quantification and Probabilistic Modeling.
ACE Group will have a strong presence at the SIAM Conference on Mathematics of Data Science in Atlanta, GA, October 21-25, including talks by me, Tomoki Koike, and Pavlos Stavrinides and posters presented by me, Dayoung Kang, and Tomoki. As a member of the SIAM activity group on Equity, Diversity, and Inclusion (EDI), I am also co-organizing a session with Tammy Kolda on Mathematical and Statistical Methods for Promoting Fairness and Equity in Algorithmic Decision-making on Friday afternoon.
Recent news
September 2024: New preprint on The Fundamental Subspaces of Ensemble Kalman Inversion (EKI) is online. EKI methods are a family of adjoint/derivative-free iterative methods for solving least-squares problems. This joint work with Christopher Beattie (Virginia Tech) provides a new analysis of EKI for solving linear least-squares problems that illuminates six fundamental subspaces of EKI analogous to the famous four fundamental subspaces of linear algebra (à la Strang).
August 2024: A warm welcome to several new group members: GT PhD students Atticus Rex and Weiting Yi, Fulbright visiting student Josie König, and GT undergraduate researchers Kashvi Mundra and Andy Yu!