Talk: Data-efficient surrogate models for tokamak digital twins

• Lorenzo Zanisi (AI/ML)

Abstract: Neural surrogate models of physics simulators are emerging ubiquitously in the Fusion community to satisfy the pressing need of fast optimisation tasks and flight simulator applications. However, gathering the training sets for these surrogates can be very expensive, an issue that is particularly crucial for high-fidelity simulators, and storing the data long-term may be impossible. In this talk I will demonstrate methodologies to obtain performing surrogate models at a lower cost in terms of training data, with applications to 0D and 2D datasets and to a reactor-relevant streaming scenario.

Discussion

• Lorenzo Zanisi (AI/ML)