Workshop II: Learning Models from Data for Multi-Fidelity Fusion Plasma Physics - IPAM

Workshop II: Learning Models from Data for Multi-Fidelity Fusion Plasma Physics

Part of the Long Program Multi-Fidelity Methods for Fusion Energy
April 13 - 17, 2026

Schedule

All times in this Schedule are Pacific Time (PT)

Monday, April 13, 2026

Morning Session

08:00-08:55
Check-In/Breakfast (Hosted by IPAM)
08:55-09:00
Welcome and Opening Remarks
Session Chair: Ionut-Gabriel Farcas
09:00-09:45
Steve Cowley (Princeton Plasma Physics Lab)
Multi-scale Plasma Turbulence — the key to fusion power
10:00-10:15
Break
10:15-11:00
Li Wang (University of Minnesota, Twin Cities)
Suppressing Plasma Instability Through Constrained Optimization
11:15-11:30
Break
11:30-12:15
Jingmei Qiu (University of Delaware)
Sampling-Based Adaptive Rank Integrators for Multi-scale Kinetic Models

Afternoon Session

12:30-14:30
Lunch (on your own)
Session Chair: Boris Kramer
14:30-15:15
Nikola Kovachki (Nvidia Corporation)
Demystifying Data-Driven Probabilistic Medium-Range Weather Forecasting
15:30-16:15
Break & Walk to CNSI
16:30-17:30
Karen Willcox (University of Texas at Austin)
Green Family Lecture #1: The Important Role of Mathematics in the Digital Twin Revolution (at UCLA CNSI Auditorium)
17:45-18:30
Reception (Location: IPAM Lobby)

Tuesday, April 14, 2026

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
Session Chair: Jingmei Qiu
09:00-09:45
Peter Benner (Max Planck Institute for Dynamics of Complex Technical Systems, Magdeburg)
Discovery and Model Reduction of Hamiltonian Systems
10:00-10:15
Break
10:15-11:00
Luis Chacon (Los Alamos National Laboratory)
Data-driven closure learning for radiation transport and plasmas
11:15-11:30
Break
11:30-12:15
Qin Li (University of Wisconsin-Madison)
Stabilizing plasma instabilities cast as PDE-constrained optimization

Afternoon Session

12:30-14:30
Lunch (on your own)
Session Chair: Cristina Rea
14:30-15:15
Jiequn Han (Flatiron Institute)
Learning Evolution Operators Across PDE Systems: Meta-Learning and Test-Time Generalization
15:30-16:15
Break & Walk to Royce Hall
16:30-17:30
Karen Willcox (University of Texas at Austin)
Green Family Lecture #2: Learning Structure-exploiting Reduced Models with Operator Inference (at UCLA Royce Hall)

Wednesday, April 15, 2026

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
Session Chair: Qin Li
09:00-09:45
Karen Willcox (University of Texas at Austin)
Multifidelity Proper Orthogonal Decomposition
10:00-10:15
Break
10:15-11:00
Diego Del-Castillo-Negrete (University of Texas at Austin)
Generative Artificial Intelligence methods for turbulence and kinetic computations
11:15-11:30
Break
11:30-12:15
Cristina Rea (Massachusetts Institute of Technology and Arizona State University)
Data-driven learning for disruption prevention and performance optimization

Afternoon Session

12:30-12:35
Group Photo
12:35-14:30
Lunch (on your own)
Session Chair: Ammar Hakim
14:30-15:15
Leonardo Zepeda-Núñez (Google Inc.)
Learning Turbulent and Chaotic Dynamics via Data-Driven Probabilistic Models
15:30-15:45
Lightning Poster Round
15:45-17:00
Poster Session & Reception (Hosted by IPAM)

Thursday, April 16, 2026

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
Session Chair: Diego Del-Castillo Negrete
09:00-09:45
Ammar Hakim (Princeton Plasma Physics Lab)
BEACONS and BEACONS-FM: Lighting a Path to Modular, Composable, Formally Verified Fusion Foundation Models
10:00-10:15
Break
10:15-11:00
Boris Kramer (University of California San Diego)
Structure-preserving Lift & Learn: Scientific machine learning for nonlinear conservative partial differential equations
11:15-11:30
Break
11:30-12:15
Benjamin Sanderse (Centrum Wiskunde & Informatica (CWI))
Structure-preserving SciML for discovering ODEs and SDEs in fluid flows

Afternoon Session

12:30-14:30
Lunch (on your own)
Session Chair: Jingwei Hu
14:30-15:15
Luis Oliveira e Silva (Instituto Superior Tecnico, University of Lisbon)
Ab initio PIC simulations and data-driven techniques for multi-fidelity or reduced plasma models
15:30-16:00
Break
16:00-16:45
Guannan Zhang (Oak Ridge National Laboratory)
Generative diffusion models for learning stochastic flow maps in particle-based kinetic simulation

Friday, April 17, 2026

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
Session Chair: Benjamin Peherstorfer
09:00-09:45
Jingwei Hu (University of Washington)
A Neural Score-Based Particle Method for the Vlasov-Maxwell-Landau System
10:00-10:15
Break
10:15-11:00
Alexander Velberg (Massachusetts Institute of Technology)
Data-driven discovery of sub-grid models for collisionless magnetic reconnection
11:15-11:30
Break
11:30-12:15
Shantenu Jha (Princeton Plasma Physics Lab)
All (Foundation) Models are wrong. Some (Foundation) Models can be made useful
Printable Version