Program

Thursday, April 6

07:00-07:50 Registration - Ronald Tutor Campus Center (TCC), room 227
07:50-08:00 Welcome
Session 1 Chair: Alireza Doostan
08:00-08:25 Houman Owhadi, California Institute of Technology
On solving/learing nonlinear PDEs with GPS
08:25-08:40 Parisa Khodabakhshi, Lehigh University
Non-intrusive parametric model reduction via data-drive operator inference
08:40-08:55 Ionut Farcas, The University of Texas at Austin
Parametric non-intrusive reduced-order models via operator inference for large-scale rotating detonation engine simulations
08:55-09:10 Petros Koumoutsakos, Harvard University
Flow reconstruction by multiresolution optimization of a discrete loss with automatic differentiation
09:10-09:25 Matthiew Darcy, California Institute of Technology
Benchmark operator learning with simple and interpretable kernal methods
09:25-09:45 Break
Session 2 Chair: Paris Perdikaris
09:45-10:10 Jacqueline Wentz, University of Colorado Boulder
Derivative-based SINDy (DSINDy): Addressing the challenge of discovering governing equations from noisy data
10:10-10:25 Anthony Gruber, Sandia National Laboratories
Canonical and noncanonical Hamiltonian model reduction through benchmark examples
10:25-10:40 Andrew Glaws, National Renewable Energy Laboratory
Scientific machine learning benchmarks to advance wind energy
10:40-10:55 Emma Lejeune, Boston Univesity
Open access benchmark datasets for assessing deep learning model accuracy and calibration in mechanics
10:55-11:10 Nicholas Nelsen, California Institute of Technology
Theory-based principles for benchmarking operator learning
11:10-12:45 Lunch (on your own)
12:45-13:45 Discussion 1
13:45-14:00 Break
Session 3 Chair: Jian-Xun Wang
14:00-14:25 Alex Gorodetsky, University of Michigan
Methods and approaches to benchmark data-driven modeling in the sparse and noisy data regimes
14:25-14:40 Matthew Levine, California Institute of Technology
Auto-differentiable data assimilation and neural ODEs for learning latent dynamics
14:40-14:55 Lekha Patel, Sandia National Laboratories
Deep learning large scale aerosol-cloud interactions from satellite imagery
14:55-15:10 Ravi Patel, Sandia National Laboratories
Error-in-variables modelling for operator learning benchmarks
15:10-15:25 Deep Ray, University of Maryland, College Park
Variationally mimetic operator network with a consistent loss function
15:25-15:45 Break
15:45-16:45 Discussion 2

17:30-19:30 Reception at The Lab (3500 S Figueroa St)

Friday, April 7

07:30-08:00 Registration
Session  4 Chair: Assad Oberai
08:00-08:25 Ellen Kuhl, Stanford University
Constitutive artificial neural networks: Establishing benchmarks for training and validation
08:25-08:40 Krishna Garikipati, University of Michigan
Graph calculus neural networks
08:40-08:55 Somdatta Goswami, Brown University
Solving partial differential equations with machine learning
08:55-09:10 Joseph Hart, Sandia National Laboratories
Solving inverse problems via neural network flow map approximation
09:10-09:25 Dongbin Xiu,  The Ohio State University
Data driven modeling of unknown systems with deep neural networks
09:25-09:45 Break
Session 5 Chair: Paris Perdikaris
09:45-10:10 Jacob Seidman, University of Pennsylvania
Dimensionality reduction in operator learning
10:10-10:25 Nick McGreivy, Princeton University
Can ML-for-science research be trusted? A crisis of weak baselines and reporting bias for ML-for-PDE solving
10:25-10:40 Siddhartha Srivastava, University of Michigan
A framework for physics-informed reinforcement learning with applications to cell migration
10:40-10:55 Vahidullah Tac, Purdue University
Benchmarks for physics-informed data-driven hyperelasticity
10:55-11:10 Opal Issan, University of California, San Diego
Predicting solar wind streams from the inner-heliosphere to earth via shifted operator inference
11:10-12:45 Lunch (on your own)
Session 6 Chair: Alireza Doostan
12:45-13:10 Leonardo Zepeda-Núñez, Google Research and University of Wisconsin-Madison
Evolve smoothly, fit consistently: Learning smooth latent dynamics For advection-dominated systems
13:10-13:25 Saibal De, Sandia National Laboratories
Improving accuracy of stochastic collocation surrogates for ODE/PDE systems via data-driven factorization of model dynamics
13:25-13:40 Dongjin Lee, University of California, San Diego
Multifidelity risk assessments for nonlinear systems under high-dimensional dependent random variables
13:40-13:55 Pieterjan Robbe, Sandia National Laboratories
Surrogate modeling of high-dimensional response surfaces with steep gradients
13:55-14:10 Yue Yu, Lehigh University
MetaP: How to transfer your hidden physics knowledge between specimens
14:10-14:30 Break
14:30-15:30 Discussion 3
15:30-16:00 Goodbyes