Method and apparatus for creating a non-parametric, data-based function model

摘要

The invention relates to a method for determining a non-parametric, data-based function model, in particular a Gaussian process model, with training data provided, the training data containing a number of measuring points defined by one or more input variables with associated output values of at least one output variable, comprising the following steps: Selecting (S2) one or more of the measuring points as specific measuring points or adding one or more additional measuring points to the training data as specific measuring points; - Assigning (S3) a measurement uncertainty value of essentially zero to the determined measurement points; and - determining (S4) the non-parametric, data-based function model according to an algorithm which depends on the determined measuring points of the modified training data and the respectively assigned measurement uncertainty values.