no code implementations • 3 Mar 2023 • Philipp Fleck, Stephan Winkler, Michael Kommenda, Michael Affenzeller
The presented results indicate, that the different random sampling strategies do not impact the overall algorithm performance significantly, and that the guided strategies suffer from becoming stuck in local optima.
no code implementations • 20 Sep 2022 • Lukas Kammerer, Gabriel Kronberger, Michael Kommenda
Fast Function Extraction (FFX) is a deterministic algorithm for solving symbolic regression problems.
1 code implementation • 13 Jun 2022 • Bogdan Burlacu, Michael Kommenda, Gabriel Kronberger, Stephan Winkler, Michael Affenzeller
This contribution discusses the role of symbolic regression in Materials Science (MS) and offers a comprehensive overview of current methodological challenges and state-of-the-art results.
no code implementations • 28 Sep 2021 • Gabriel Kronberger, Lukas Kammerer, Bogdan Burlacu, Stephan M. Winkler, Michael Kommenda, Michael Affenzeller
In this chapter we take a closer look at the distribution of symbolic regression models generated by genetic programming in the search space.
no code implementations • 28 Sep 2021 • Lukas Kammerer, Gabriel Kronberger, Bogdan Burlacu, Stephan M. Winkler, Michael Kommenda, Michael Affenzeller
Symbolic regression is a powerful system identification technique in industrial scenarios where no prior knowledge on model structure is available.
no code implementations • 8 Sep 2021 • Philipp Fleck, Manfred Kügel, Michael Kommenda
In cases of large proportions of missing data, imputing is often not feasible, and removing observations containing missing values is often the only option.
no code implementations • 8 Sep 2021 • Florian Holzinger, Michael Kommenda
The gathered data was used for modeling (and therefore monitoring) a healthy state.
no code implementations • 1 Sep 2021 • Michael Kommenda, Johannes Karder, Andreas Beham, Bogdan Burlacu, Gabriel Kronberger, Stefan Wagner, Michael Affenzeller
In this contribution we revisit the core principles of optimization networks and demonstrate their suitability for solving machine learning problems.
no code implementations • 1 Sep 2021 • Michael Kommenda, Andreas Beham, Michael Affenzeller, Gabriel Kronberger
Multi-objective symbolic regression has the advantage that while the accuracy of the learned models is maximized, the complexity is automatically adapted and need not be specified a-priori.
no code implementations • 24 Aug 2021 • Bogdan Burlacu, Michael Affenzeller, Michael Kommenda
This paper describes a methodology for analyzing the evolutionary dynamics of genetic programming (GP) using genealogical information, diversity measures and information about the fitness variation from parent to offspring.
no code implementations • 24 Aug 2021 • Lukas Kammerer, Gabriel Kronberger, Michael Kommenda
The performance of genetic programming is compared with random forests and linear regression.
no code implementations • 3 Aug 2021 • Gabriel Kronberger, Evgeniya Kabliman, Johannes Kronsteiner, Michael Kommenda
A major drawback is the calibration of model parameters that depend on processing conditions.
4 code implementations • 29 Jul 2021 • William La Cava, Patryk Orzechowski, Bogdan Burlacu, Fabrício Olivetti de França, Marco Virgolin, Ying Jin, Michael Kommenda, Jason H. Moore
We assess 14 symbolic regression methods and 7 machine learning methods on a set of 252 diverse regression problems.
no code implementations • 19 Jul 2021 • Gabriel Kronberger, Michael Kommenda, Andreas Promberger, Falk Nickel
We have therefore used so-called factor variables for representing nominal variables in symbolic regression models.
no code implementations • 6 Jul 2021 • Gabriel Kronberger, Lukas Kammerer, Michael Kommenda
We describe a method for the identification of models for dynamical systems from observational data.
no code implementations • 29 Mar 2021 • Gabriel Kronberger, Fabricio Olivetti de França, Bogdan Burlacu, Christian Haider, Michael Kommenda
Both algorithms are able to identify models which conform to shape constraints which is not the case for the unmodified symbolic regression algorithms.
no code implementations • 3 Feb 2019 • Bogdan Burlacu, Michael Affenzeller, Gabriel Kronberger, Michael Kommenda
Diversity represents an important aspect of genetic programming, being directly correlated with search performance.
no code implementations • 23 Sep 2013 • Michael Kommenda, Gabriel Kronberger, Christoph Feilmayr, Michael Affenzeller
The approach is based on unguided symbolic regression (every variable present in the dataset is treated as the target variable in multiple regression runs) and a novel variable relevance metric for genetic programming.
no code implementations • 16 May 2013 • Gabriel Kronberger, Michael Kommenda
In the proposed approach we use a grammar for the composition of covariance functions and genetic programming to search over the space of sentences that can be derived from the grammar.
no code implementations • 10 Dec 2012 • Gabriel Kronberger, Stefan Fink, Michael Kommenda, Michael Affenzeller
In the proposed approach multiple symbolic regression runs are executed for each variable of the dataset to find potentially interesting models.