research
We focus on developing new computational methods for uncertainty quantification and scientific computing. Our work aims to design efficient, interpretable, and trustworthy algorithms for complex problems in scientific and engineering fields. To this end we use insights from numerical analysis, Bayesian statistics, and machine learning. My research interests are in
1. Uncertainty quantification for inverse problems
-
Bayesian (statistical) inverse problems, e.g., BayesianPET, 2020 and BayesianPDE, 2024
-
New prior or sampling methods, e.g., AdaptiveMCMC, 2017 and RoundtripPrior, 2024
-
Deep generative models for inverse problems, e.g., DGM-EIT, 2024
2. AI for science
-
Artificial intelligence for scientific computing, e.g., HQSNet, 2024 and DuNets-RMA, 2024
-
Artificial intelligence for bioinformatics, e.g., UQguidedScreening, 2023