Expensive Optimisation Exemplified by ECG Simulator Parameter Tuning

This article describes the tuning of an Electrocardiogram (ECG) simulator as a benchmark problem to show the application of surrogate modelling in complex global optimisation. After presenting the background on ECG, its simulation and the optimisation task, the main concepts and methods of surrogate modelling and Efficient Global Optimisation (EGO) are presented. Here, next to the standard techniques regularly involved in the algorithm, alternative approaches are discussed briefly. Finally, first results applying the depicted algorithm on the ECG simulator optimisation problem are presented.