Uncertainty Quantification Seminar Yous van Halder (CWI)
Neural networks for multifidelity surrogate modelling
- https://www.cwi.nl/research/groups/scientific-computing/events/uncertainty-quantification-seminar-yous-van-halder-cwi
- Uncertainty Quantification Seminar Yous van Halder (CWI)
- 2018-12-20T10:00:00+01:00
- 2018-12-20T11:00:00+01:00
- Neural networks for multifidelity surrogate modelling
- What Scientific Computing English
- When 20-12-2018 from 10:00 to 11:00 (Europe/Amsterdam / UTC100)
- Where CWI, L102
- Contact Name Benjamin Sanderse
- Web Visit external website
-
Add event to calendar
iCal
Neural networks for multifidelity surrogate modelling
In this talk we discuss how neural networks can be used together with multifidelity techniques to accelerate the parametric solution of time-dependent partial differential equations. Based on a small number of high-resolution (high-fidelity) simulations, a neural network is trained ('offline') that maps low-fidelity simulation outputs to achieve an accuracy similar to the high-fidelity simulations in 'online' mode. We show promising results for non-linear equations including the flow over a backward-facing step.