Inverse problems occur in a wide range of areas, such as medical imaging, astronomy and geophysics.
Our research group investigates numerous computational methods, which are used in an effort to combine modelling and measurement. In practice, the goal is to develop tools which can be used to analyse the reliability of model forecasts.
Parameter estimation, inverse problems and data assimilation are, in part, overlapping concepts, which are used in various scientific communities. Parameter estimation usually refers to the identification of a limited set of model parameters, while inverse problems often involve the estimation of continuous profiles or functions.
An excellent example of data assimilation is weather forecasting: continuous-time identification and forecasting of the state of a large dynamic system.
Inverse problem research areas
- Adaptive MCMC methods for statistical inversions.
- Data assimilation methods.
- Optimal design of experiments, stochastic optimisation methods.
- Wavelet methods.
Read more on the group wiki page.