Workshop III: Validation and Guarantees in Learning Physical Models: from Patterns to Governing Equations to Laws of Nature - IPAM

Workshop III: Validation and Guarantees in Learning Physical Models: from Patterns to Governing Equations to Laws of Nature

Part of the Long Program Machine Learning for Physics and the Physics of Learning
October 28 - November 1, 2019

Schedule

All times in this Schedule are Pacific Time (PT)

Monday, October 28, 2019

Morning Session

08:00-08:50
Check-In/Breakfast (Hosted by IPAM)
08:50-09:00
Welcome & Opening Remarks: Dean Miguel García-Garibay (Dean of Physical Sciences, UCLA) and Dima Shlyakhtenko (Director, IPAM)
09:00-09:50
Petros Koumoutsakos (Harvard University)
Machine Learning for Fluid mechanics
10:00-10:15
Break
10:15-11:05
Mohammad Farazmand (North Carolina State University)
Multiscale analysis of accelerated gradient methods in machine learning
11:15-11:30
Break
11:30-12:20
Krithika Manohar (California Institute of Technology)
Kernel Analog Forecasting: Multiscale Problems

Afternoon Session

12:30-14:30
Lunch (on your own)
14:30-15:20
Boumediene Hamzi (Imperial College)
Machine Learning and Dynamical Systems meet in Reproducing Kernel Hilbert Spaces
15:30-15:45
Break
15:45-16:35
Mauro Maggioni (Johns Hopkins University)
Title: Learning Interaction laws in agent-based systems
16:45-17:15
Lightning Poster Presentations
17:15-18:30
Poster Session & Reception (Hosted by IPAM)

Tuesday, October 29, 2019

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
09:00-09:50
Steve Brunton (University of Washington)
Data-driven discovery and control of nonlinear flow physics
10:00-10:15
Break
10:15-11:05
Jean-Christophe Loiseau (École Nationale Supérieure d'Arts et Métiers)
A brief overview of SINDy
11:15-11:30
Break
11:30-12:20
Jens Eisert (Freie Universität Berlin)
Learning classical and quantum dynamical laws from data

Afternoon Session

12:30-14:30
Lunch (on your own)
14:30-15:20
Paris Perdikaris (University of Pennsylvania)
Overcoming gradient pathologies in constrained neural networks
15:30-16:00
Break
16:00-16:50
Maziar Raissi (Brown University)
Hidden Physics Models: Machine Learning of Non-Linear Partial Differential Equations

Wednesday, October 30, 2019

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
09:00-09:50
Klaus-Robert Müller (Technische Universität Berlin)
Understanding ML models
10:00-10:15
Break
10:15-11:05
Bjoern Baumeier (Technische Universiteit Eindhoven)
Towards Describing Macroscopic Behavior of Electronic Processes from ML-enhanced Multiscale Models
11:15-11:30
Break
11:30-12:20
Samory Kpotufe (Princeton University)
Some new Insights On Transfer Learning

Afternoon Session

12:30-14:30
Lunch (on your own)
14:30-15:20
Gitta Kutyniok (Ludwig-Maximilians-Universität München)
An Information Theoretic Approach to Validate Deep Learning-Based Algorithms
15:30-16:00
Break
16:00-16:50
John Maddocks (École Polytechnique Fédérale de Lausanne (EPFL))
The cgDNA sequence-dependent coarse-grain model of dsDNA: Bridging the scales from Molecular Dynamics to Bioinformatics

Thursday, October 31, 2019

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
09:00-09:50
Christof Schuette (Freie Universität Berlin)
Learning reduced models in molecular dynamics: From finding transition manifolds to effective dynamics
10:00-10:15
Break
10:15-11:05
Marina Meila (University of Waterloo)
Validation and Reproducibility by Geometry, for Unsupervised Learning
11:15-11:30
Break
11:30-12:20
Jelena Bradic (University of California, San Diego (UCSD))
Causality and machine learning for scientific discovery

Afternoon Session

12:30-14:30
Lunch (on your own)
14:30-15:20
Aaditya Ramdas (Carnegie Mellon University)
Doubly-sequential experimentation
15:30-16:00
Break
16:00-17:00
Discussion

Friday, November 1, 2019

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
09:00-09:50
Bing Brunton (University of Washington)
Fields and flows in neural dynamics
10:00-10:15
Break
10:15-11:05
Stefan Klus (Freie Universität Berlin)
Data-driven transfer operator approximation, model reduction, and system identification
11:15-11:30
Break
11:30-12:20
Denis Zorin (New York University)
Machine learning-based surrogates for optimization with physical constraints

Afternoon Session

12:30-14:30
Lunch (on your own)
14:30-15:20
Joan Bruna (New York University)
Geometric Insights into the Convergence of Nonlinear TD Learning
15:30-16:00
Break
16:00-16:50
Lukas Pospisil (Technical University of Ostrava (VŠB))
Towards a low-cost scalable data-driven modeling of coupled systems and non-stationary time series analysis
17:00-17:00
Conclusion
Printable Version