Joseph Lee Hart

Scientific Machine Learning

Author profile picture

Scientific Machine Learning

joshart@sandia.gov

Sandia National Laboratories, New Mexico
P.O. Box 5800
Albuquerque, NM 87185-1327

Biography

Joseph is interested in quantifying, prioritzing, and mitigating uncertainty in large-scale optimization problems constrained by differential equations. His research spans numerical optimization, sensitivity analysis, inverse problems, optimal experimental design, and scientific machine learning in the service of outer loop analysis. Joseph focuses on finding and exploiting low dimensional structure which arises from taking a wholestic perspective on the scientific computing pipeline from model development to decision-making.

Education

Joseph earned a B.S. in Mathematics from North Carolina State University in 2014, a M.S. in Applied Mathematics from North Carolina State University in 2016, and a Ph.D. in Applied Mathematics with an Interdisciplinary Track in Statistics in 2018, with his dissertation "Extensions of Global Sensitivity Analysis: Theory, Computation, and Application."

Publications

Joseph Lee Hart, Bart G van Bloemen Waanders, (2022). Characterizing and Propagating Model Discrepancy in PDE-Constrained Optimization USACM Thematic Conference on Uncertainty Quantification for Machine Learning Integrated Physics Modeling Document ID: 1595711

Joseph Lee Hart, Mamikon Gulian, Indu Manickam, Laura Painton Swiler, (2022). Facilitating Atmospheric Source Inversion via Operator Regression SIAM Conference on Mathematics of Planet Earth Document ID: 1563153

Mamikon Gulian, Joseph Lee Hart, Indu Manickam, Laura Painton Swiler, (2022). Facilitating Atmospheric Source Inversion via Deep Operator Network Surrogates Esco 2022 Document ID: 1550971

Joseph Lee Hart, Bart G van Bloemen Waanders, (2022). Hyper-differential sensitivity analysis with respect to model discrepancy ASCR RISE Project Meeting Document ID: 1528576

Joseph Lee Hart, Bart G van Bloemen Waanders, (2022). Hyper-differential sensitivity analysis with respect to model discrepancy ECCOMAS Congress Document ID: 1528577

Joseph Lee Hart, (2022). Characterizing and propagating model discrepancy in PDE-constrained optimization USACM Thematic Conference on Uncertainty Quantification for Machine Learning Integrated Physics Modeling (MLIP) Document ID: 1494158

Joseph Lee Hart, (2022). Enabling and interpreting hyper-differential sensitivity analysis for Bayesian inverse problems SIAM Conference on Uncertainty Quantification Document ID: 1493904

Joseph Lee Hart, (2022). Facilitating atmospheric source inversion via operator regression SIAM Conference on Mathematics of Planet Earth Document ID: 1438195

Isaac Paul Sunseri, Joseph Lee Hart, Alen Alexanderian, Bart G van Bloemen Waanders, (2021). Hyper-Differential Sensitivity Analysis of Inverse Problems Governed by PDEs Joint Mathematics Meetings 2022 Document ID: 1405614

Joseph Lee Hart, Bart G van Bloemen Waanders, (2021). Hyper-differential sensitivity analysis with respect to model discrepancy The 8th European Congress on Computational Methods in Applied Sciences and Engineering Document ID: 1393173

Joseph Lee Hart, (2021). Hyper-differential sensitivity analysis in PDE-constrained inverse problems UC Merced Virtual Seminar Talk Document ID: 1392621

Joseph Lee Hart, Bart G van Bloemen Waanders, (2021). Hyper-differential sensitivity analysis for model form error Mechanistic Machine Learning and Digital Twins for Computational Science, Engineering & Technology Document ID: 1367680

William Mcneil Reese, Joseph Lee Hart, Bart G van Bloemen Waanders, Mauro Perergo, John Davis Jakeman, Arvind Saibaba, (2021). Bedrock Inversion and Hyper Differential Sensitivity Analysis for the Shallow Ice Model Numerical Analysis in Data Science Transition Workshop Document ID: 1317435

Joseph Lee Hart, (2021). Enabling and interpreting hyper-differential sensitivity analysis for Bayesian inverse problems Project meeting with university collaborators on the ASCR RISE project https://www.osti.gov/search/identifier:1865768 Document ID: 1305113

Joseph Lee Hart, (2021). Hyper-differential sensitivity analysis of model form error Mechanistic Machine Learning and Digital Twins for Computational Science conference Document ID: 1292007

Joseph Lee Hart, (2021). Hyper-Differential Sensitivity Analysis for Robust Machine Learned Surrogate Models SIAM Conference on Computational Science and Engineering https://www.osti.gov/search/identifier:1847611 Document ID: 1268799

Bart G van Bloemen Waanders, Joseph Lee Hart, (2020). Hyper-Differential Sensitivity Analysis: Managing High Dimensional Uncertainty in Large-Scale Optimization Problems Uncertainty Management and Machine Learning in Engineering Applications https://www.osti.gov/search/identifier:1877814 Document ID: 1232146

Elizabeth Newman, Lars Ruthotto, Joseph Lee Hart, Bart G van Bloemen Waanders, (2020). Train Like a (Var)Pro: Efficient Training of Neural Networks with Variable Projection SIAM Journal on Mathematics of Data Science https://www.osti.gov/search/identifier:1834344 Document ID: 1184550

Joseph Lee Hart, (2020). Hyper-Differential Sensitivity Analysis: Managing High Dimensional Uncertainty in Large-Scale Optimization Numerical Analysis Seminar at North Carolina State University (Conducted online) https://www.osti.gov/search/identifier:1773277 Document ID: 1115024

Joseph Lee Hart, (2020). Hyper-Differential Sensitivity Analysis: Managing High Dimensional Uncertainty in Large-Scale Optimization Numerical Analysis Seminar at North Carolina State University (Online due to COVID-19) Document ID: 1104839

Joseph Lee Hart, Steven Gilmore, Pierre Gremaud, Christian Olsen, Jesper Mehlsen, Mette Olufsen, (2020). Classification of autonomic dysfunction through data analytics Mathematical Biosciences and Engineering https://www.osti.gov/search/identifier:1834334 Document ID: 1103307

Joseph Lee Hart, Pierre Gremaud, (2020). Robustness of Sobol? indices to distributional uncertainty SIAM Conference on Uncertainty Quantification https://www.osti.gov/search/identifier:1768131 Document ID: 1102890

Isaac Paul Sunseri, Bart G van Bloemen Waanders, Joseph Lee Hart, Alen Alexandarian, (2020). Hyper-differential sensitivity analysis of PDE-constrained inverse problems AWM Math Research Competition 2020 https://www.osti.gov/search/identifier:1767902 Document ID: 1092566

Elizabeth – x Newman, Lar – x Ruthotto, Joseph Lee Hart, Bart G van Bloemen Waanders, Bart G van Bloemen Waanders, (2020). Efficient Training of Neural Network Surrogate Methods with Variable Projection Mathematical and Scientific Machine Learning https://www.osti.gov/search/identifier:1767098 Document ID: 1067134

Joseph Lee Hart, Bart G van Bloemen Waanders, Arvind Saibaba, (2020). Randomized Algorithms for Generalized Singular Value Decomposition with Application to Sensitivity Analysis Numerical Linear Algebra with Applications https://www.osti.gov/search/identifier:1769894 Document ID: 1091454

Isaac Paul Sunseri, Joseph Lee Hart, Alen Alexandarian, Bart G van Bloemen Waanders, (2020). Quantifying the relative importance of complimentary parameters in PDE-based inverse problems NSFResearch Training Group Meeting at NC State University https://www.osti.gov/search/identifier:1763861 Document ID: 1079641

Joseph Lee Hart, (2019). Hyper-Differential Sensitivity Analysis to Support Geophysical Inverse Problems AGU Fall Meeting https://www.osti.gov/search/identifier:1643269 Document ID: 1055426

Charles Vollmer, Alex Fout, Giorgia Bettin, Edward N Matteo, Paul Conrad Schwering, Christian Poppeliers, Gungor Didem Beskardes, Joseph Lee Hart, Bart G van Bloemen Waanders, Chester J Weiss, David John Stracuzzi, Avery Ted Cashion, (2019). Event Detection and Characterization in a Network of Subsurface Sensors American Geophysical Union (AGU) Fall Meeting 2019 Document ID: 997583

Joseph Lee Hart, (2019). Robustness of Sobol’ indices to distributional uncertainty SIAM Conference on Uncertainty Quantification Document ID: 997146

Joseph Lee Hart, (2019). Global Sensitivity Analysis for PDE-constrained Optimization SIAM CS&E Document ID: 997294

Joseph Lee Hart, (2019). Hyper-differential sensitivity analysis for PDE-constrained optimization: Methods and software International Congress on Industrial and Applied Mathematics Document ID: 997295

Joseph Lee Hart, (2019). Hyper-differential sensitivity analysis to support geophysical inverse problems AGU Fall Meeting 2019 Document ID: 997144

Joseph Lee Hart, (2019). Hyper-differential sensitivity analysis for PDE-constrained optimization: Methods and software International Congress on Industrial and Applied Mathematics https://www.osti.gov/search/identifier:1641004 Document ID: 984557

Joseph Lee Hart, Bart G van Bloemen Waanders, (2019). Global Sensitivity Analysis for PDE-constrained Optimization SIAM Conference on Computational Science and Engineering https://www.osti.gov/search/identifier:1601533 Document ID: 914693

Joseph Lee Hart, Bart G van Bloemen Waanders, (2019). Computationally Efficient Parameter Sensitivity Analysis for PDE-Constrained Optimization SIAM Conference on Computational Science and Engineering https://www.osti.gov/search/identifier:1601532 Document ID: 914694

Joseph Lee Hart, (2018). Computationally Efficient Parameter Sensitivity Analysis for PDE-Constrained Optimization SIAM Conference on Computational Science and Engineering https://www.osti.gov/search/identifier:1806909 Document ID: 889197

Bart G van Bloemen Waanders, Joseph Lee Hart, (2018). Sensitivity Analysis for Optimal Control The Minerals, Metals, and Materials Society Annual Conference 2018 https://www.osti.gov/search/identifier:1501995 Document ID: 772303

Jordan Massad, Erin Acquesta, Joseph Lee Hart, (2017). Mathematics and Statistics at Sandia National Laboratories: Perspectives from an Engineering Scientist, Analyst, and Student Intern NCSU Mathematics and Statistics Recruiting https://www.osti.gov/search/identifier:1511133 Document ID: 726091

Jordan Massad, Jordan Massad, Erin Acquesta, Erin Acquesta, Joseph Lee Hart, Joseph Lee Hart, (2017). Mathematics and Statistics at Sandia National Laboratories: Perspectives from an Engineering Scientist, Analyst, and Student Intern NCSU Mathematics and Statistics Recruiting Document ID: 725416

Showing Results. Show More Publications