$\vec{w}h\alpha\mathfrak{t}\;\; \forall\mathbb{R}\varepsilon\ldots$

PINNs?


Who?
Marwa Zainelabdeen (WIAS)
When?
2024/06/28, 13:00 s.t.
Before the MATH+ Friday colloquium talk by María Ángeles García-Ferrero
Where?
Zuse Institute Berlin (ZIB)
About what?

In convection-dominated regimes, solutions of convection-diffusion problems often have steep gradients. Traditional numerical methods struggle with these issues. Recently, physics-informed neural networks (PINNs) have emerged, minimizing residuals at collocation points. This talk introduces specialized loss functionals for PINNs tailored to these problems.