Making Sense of Large Computational Experiments

Fredrik Wrede
Division of Scientific Computing
Department of Information Technology
Uppsala University
Uppsala

Abstract:

Large computational experiments involving parameter sweep applications (PSAs), where a simulator acts much like a "black box", can be used to for e.g. robustness analysis of underlying models, uncertainty quantification, computational design and model exploration. However, for complex models involving many parameter and with little a priori knowledge, such sweeps can become massive. Thus, making it impossible to manually analyze the results and organize the data.

In this talk, I will give a short introduction to the problems at hand and how we might use machine learning techniques and e-science tools for automatic and efficient scientific discovery of sweeps supporting temporal data derived from both deterministic and stochastic models.