Tian Yu Yen

Scientific Machine Learning

Scientific Machine Learning

tyen@sandia.gov

Sandia National Laboratories, New Mexico
P.O. Box 5800
Albuquerque, NM 87185-1318

Biography

Tian Yu joined Sandia as a postdoc in 2021. His current research focuses on methods of aleotoric and epistemic uncertainty quantification for inverse problems as well as techniques for quantifiying uncertainty in machine learning algorithms. He has a background is in measure-theoretic inversion, Bayesian inference, and statistical density estimation. His personal vision and mission statement is included below.

"Through my career in mathematics, I aim to train and mentor the next generation of STEM professionals–especially those individuals who have faced systematic and historic barriers to the field. By pursuing impactful research and demonstrating inspiring leadership, I work to make rigorous mathematical thinking and advanced statistical tools accessible to all who seek them."

Education

PhD, Applied Mathematics, University of Colorado Denver, July 2021

MS, Statistics, University of Colorado Denver, December 2018

BA, Psychology, Reed College, May 2009

Publications

Tian Yu NMN Yen, Timothy Michael Wildey, (2022). Using Manifold Learning to Enable Computationally Efficient Stochastic Inversion with High-dimensional Data Wccm-apcom 2022 Document ID: 1573846

Timothy Michael Wildey, Gianluca Geraci, Michael S. Eldred, John Davis Jakeman, Owen Davis, Teresa Portone, Tian Yu NMN Yen, Bryan William Reuter, Alex Gorodetsky, Ahmad Rushdi, Daniele Schiavazzi, Lauren Partin, (2022). Embedded uncertainty estimation for data-driven surrogates to enable trustworthy ML for UQ Machine Learning and Deep Learning Conference Document ID: 1573362

Timothy Michael Wildey, Troy Butler, Tian Yu NMN Yen, (2022). A Probabilistic Characterization of Aleatoric and Epistemic Uncertainty in Solutions to Stochastic Inverse Problems Using Machine Learning Surrogate Models SIAM Conference on Uncertainty Quantification Document ID: 1504576

Tian Yu NMN Yen, Timothy Michael Wildey, Troy Butler, (2022). Quantifying Aleatoric and Epistemic Uncertainties in RLC Circuits with Data-consistent Inversion SIAM Conference on Uncertainty Quantification Document ID: 1504683

Tian Yu NMN Yen, Timothy Michael Wildey, (2021). Using Manifold Learning to Enable Computationally Efficient Stochastic Inversion with High-dimensional Data 15th World Congress on Computational Mechanics Document ID: 1392127

Showing Results. Show More Publications