Symbolic Regression

Explore the power of symbolic regression for equation learning!


What is Symbolic Regression?

Symbolic regression is a supervised learning task where the goal is to find equations that fit data (equation learning). Symbolic regression models allow to predict one or multiple variables from known variables. In contrast to other regression methods, the task is not only to identify fitting parameter values for a fixed equation structure, but instead to find the complete equation including fitting parameter values.
Symbolic regression was coined by John Koza in the context of genetic programming (GP). GP is an evolutionary algorithm for symbolic regression. It manages a set of equations (population) and recombines parts from well-fitting equations to produce new equations. This processes is repeated over many generations to produce better and better solutions starting from a set of randomly initialiezd equations.

Many different algorithms for symbolic regression have been described as alternatives to GP. One approach that can be useful to find short equations is systematic enumeration of the set of equations as in Grammar Enumeration or Exhaustive Symbolic Regression.

Our projects

AstroSymReg - Accelerating the Physical Sciences with Symbolic Regression

We develop and apply symbolic regression algorithms to create models for astrophysics, such as emulators for the linear and the nonlinear mass power spectrum.
Project duration: 2024 - ongoing

ProMetHEus - Production and processing of metals for high-performance, energy efficiency, environmental protection and sustainability

The project, lead by LKR Light Metal Competence Center Ranshofen, Austrian Institute of Technology (AIT), supports companies in the materials processing industries to produce sustainably and efficiently. We develop symbolic regression algorithms and models for new process routes.
Project duration: 2024 - ongoing

TransMet - Fundamentals and tools for engineering of high quality recycled and CO2 reduced strip steels

We develop algorithms for the adaptation of material models in this project which is lead by the Materials Center Leoben. Focus of our activities are the combination of physics-based models with data-driven models using symbolic regression.
Project duration: 2021 - 2024

Josef Ressel Center for Symbolic Regression

Within the Josef Ressel Centre for Symbolic Regression we developed new symbolic regression algorithms as well as a methodological and technical framework for incremental model adaptation for handling concept drift. We used symbolic regression for modelling components of powertrains, friction systems, and plastics recycling plants.
Project duration: 2018 - 2022


FH-Prof. DI Dr. Gabriel Kronberger

Professor for Business Intelligence and Data Engineering
Heuristic and Evolutionary Algorithms Laboratory (HEAL)
University of Applied Sciences Upper Austria (FH OÖ)
Phone: +43 80484 22320