Research Area B
We will develop tools to estimate and handle the uncertainty in data-driven scientific models and algorithms, and exploit this information for experimental design.
Machine learning needs to improve the methods by which uncertainty can be quantified and handled in large and complex scientific models. This includes techniques to separate uncertainty about data, predictions, parameters and models. We will therefore develop tools to estimate and handle the uncertainty in complex scientific models and algorithms and exploit this information for experimental design. This includes research on the following themes:
- Uncertainty quantification, propagation and control for complex models
- Uncertainty calibration and model selection
- Bayesian active learning for experimental design