Skip to main content
  • Expression of Interest

    Josef Ressel Center for Symbolic Regression

    We are developing algorithms for symbolic regression. More information
    The key characteristic of symbolic regression is that it produces models as a closed-form mathematical expression (formula) which is open to interpretation and can be integrated easily into other software systems. The input is a dataset as well as the set of operators and functions that are allowed to be used in the formula. It works well in settings with a small number of numerical variables (5 - 100) and non-linear dependencies for instance for engineering models.  Our group has extensive experience in application of SR in industrial projects and is capable to develop custom SR algorithms for specific problems.