Workshop IV: Multi-Fidelity Methods to Enable Robust Optimization and Real-Time Control of Fusion Processes - IPAM

Workshop IV: Multi-Fidelity Methods to Enable Robust Optimization and Real-Time Control of Fusion Processes

Part of the Long Program Multi-Fidelity Methods for Fusion Energy
May 18 - 22, 2026

Schedule

All times in this Schedule are Pacific Time (PT)

Monday, May 18, 2026

Morning Session

08:00-08:55
Check-In/Breakfast (Hosted by IPAM)
08:55-09:00
Welcome and Opening Remarks
SESSION CHAIR: Monday AM - TBA
09:00-09:45
Mark Kostuk (General Atomics)
Mixed-Fidelity, Multi-Model Optimization within the DIII-D Digital Twin
10:00-10:15
Break
10:15-11:00
Florian Schaefer (New York University)
Information geometric regularization for sensitivities of flows with shocks
11:15-11:30
Break
11:30-12:15
Pablo Rodriguez-Fernandez (Massachusetts Institute of Technology)
Accelerating core plasma design and optimization with surrogate-assisted transport workflows

Afternoon Session

12:30-14:30
Lunch (on your own)
SESSION CHAIR: Monday PM - TBA
14:30-15:15
Bethany Lusch (Argonne National Laboratory)
Scalable graph-based surrogate models for unstructured meshes
15:30-15:45
Lightning Poster Round
15:45-17:00
Poster Session & Reception (Hosted by IPAM)

Tuesday, May 19, 2026

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
SESSION CHAIR: TUESDAY AM - TBA
09:00-09:45
Jeff Schneider (Carnegie Mellon University)
Reinforcement Learning and Bayesian Optimization for Nuclear Fusion
10:00-10:15
Break
10:15-11:00
Michael Abdelmalik (Technische Universiteit Eindhoven)
"A Multi-Fidelity Framework for Rarefied Dynamics: Hierarchical Models, Multi-Scale Closures and Acceleration with Neural Green's Operators"
11:15-11:30
Break
11:30-12:15
Cory Hauck (Oak Ridge National Laboratory)
Approximate entropy-based moment closures

Afternoon Session

12:30-14:30
Lunch (on your own)
SESSION CHAIR: TUESDAY PM - TBA
14:30-15:30
Discussion
15:30-16:00
Break
16:00-17:00
Discussion

Wednesday, May 20, 2026

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
SESSION CHAIR: WEDNESDAY AM - TBA
09:00-09:45
Brian Spears (Lawrence Livermore National Laboratory)
The Genesis Mission : A nation-scale AI effort for Mission-Driven Science, National Security, and Applied Energy
10:00-10:15
Break
10:15-11:00
Yunan Yang (Cornell University)
Adjoint Direct Simulation Monte Carlo (DSMC) for Optimization and Control of Rarefied Flows
11:15-11:30
Break
11:30-12:15
Rebekah White (Sandia National Laboratories)
Surrogate-Based Optimization of Magnetized Liner Inertial Fusion (MagLIF) Target Design

Afternoon Session

12:30-12:40
Group Photo
12:40-14:30
Lunch (on your own)
SESSION CHAIR: WEDNESDAY PM - TBA
14:30-15:30
Discussion
15:30-16:00
Break
16:00-17:00
Discussion

Thursday, May 21, 2026

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
SESSION CHAIR: THURSDAY AM - TBA
09:00-09:45
Gianluca Geraci (Sandia National Laboratories)
Recent Advancements in Multi-Fidelity Approaches
10:00-10:15
Break
10:15-11:00
Andy Rothstein (Princeton University)
AI-Driven Event Prediction and Fault-Resilient Plasma Control on DIII-D
11:15-11:30
Break
11:30-12:15
Sophia Henneberg (Massachusetts Institute of Technology)
Optimization of Quasi-Axisymmetric Stellarator–Tokamak Hybrids

Afternoon Session

12:30-14:30
Lunch (on your own)
SESSION CHAIR: THURSDAY PM - TBA
14:30-15:15
Cosmas Heiss (École Polytechnique Fédérale de Lausanne (EPFL))
Developments in Magnetic Control at TCV and Achievement of Stabilized Doublet Plasmas
15:30-16:00
Break
16:00-17:00
Discussion

Friday, May 22, 2026

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
09:00-09:45
Stephen Becker (University of Colorado Boulder)
Some new bi-fidelity methods for UQ and optimization
10:00-10:15
Break
10:15-11:00
Simon Van Mulders (EPFL (Ecole Polytechnique Fédérale de Lausanne))
Model-aided TCV operation through routine pre-shot, real-time and post-shot self-consistent transport and equilibrium simulations, towards reactor-relevant tokamak plasma scenario optimization and control
11:15-11:30
Break
11:30-12:15
Shancong Mou (University of Minnesota, Twin Cities)
Derivative-Informed Training of Neural Operators on the Fly
Printable Version