Ulteriori informazioni
Symbolic regression (SR) is one of the most powerful machine learning techniques that produces transparent models, searching the space of mathematical expressions for a model that represents the relationship between the predictors and the dependent variable without the need of taking assumptions about the model structure.
Sommario
Contents
Preface
Symbols and Notation
1. Introduction
2. Basics of Supervised Learning
3. Basics of Symbolic Regression
4. Evolutionary Computation and Genetic Programming
5. Model Validation, Inspection, Simplification and Selection
6. Advanced Techniques
7. Examples and Applications
8. Conclusion
Appendix
Bibliography
Info autore
The authors are all affiliated with the University of Applied Sciences (UAS) Upper Austria.
Gabriel Kronberger is professor for data engineering and business intelligence. His research interests are symbolic regression and machine learning as well as probabilistic graphical models.
Bogdan Burlacu is a research assistant. His main focus is the study of genetic programming evolutionary dynamics in symbolic regression scenarios.
Michael Kommenda is a research assistant. He has been applying symbolic regression methods in various industrial projects and application scenarios.
Stephan M. Winkler is professor for medical and bioinformatics and head of the bioinformatics research group. His research interests despite bioinformatics include genetic programming, nonlinear model identification and machine learning.
Michael Affenzeller is professor for heuristic optimization and machine learning and head of the Heuristic and Evolutionary Algorithms Laboratory. Furthermore, he is the vice dean for research and overall head of the COMET project for heuristic optimization in production and logistics (HOPL).
Riassunto
Symbolic regression (SR) is one of the most powerful machine learning techniques that produces transparent models, searching the space of mathematical expressions for a model that represents the relationship between the predictors and the dependent variable without the need of taking assumptions about the model structure.
