From machine-learned preconditioners to control-oriented models
A good understanding of your problem is essential when applying machine learning tools.
We know this, because our team has developed cutting-edge tools, such as
- Data-driven preconditioners (PreconNet), applied to plasma simulations but applicable in a wider context
- Surrogate models for tokamak turbulence (QLKNN)
- Machine-learned solver selection for Finite Element (FEM) codes.
- Surrogate models for solution prediction in Finite Element codes.