A Subjective Review of the State of the Art in Model - Based Parameter Tuning

Over recent years, the model-based parameter tuning of computational systems has become an emergent research topic. When the considered systems are subject of stochastic effects, the parameter tuning can be seen as a noisy optimization task, as often faced in production engineering. In this paper, the current research on parameter optimization is discussed based on experiences from an engineering background. An extended framework of important steps in a model-based parametertuning procedure is proposed which enhances known frameworks with data transformation and model validation steps. For each step, references to important literature are provided and topics for future research are identified.