Michael Kommenda, Andreas Beham, Michael Affenzeller, Gabriel Kronberger,
"Complexity Measures for Multi-Objective Symbolic Regression"
, in R. Moreno-Diaz, F.Pichler, A. Quesada-Arencibia: Computer Aided Systems Theory ? EUROCAST 2015. 15th International Conference, Las Palmas de Gran Canaria, Spain, February 8-13, 2015, Revised Selected Papers, Serie Lecture Notes in Computer Science (LNCS), Vol. 9520, Springer, Seite(n) 409-416, 2015, ISBN: 978-3-319-27339-6, ISSN: 0302-9743
Original Titel:
Complexity Measures for Multi-Objective Symbolic Regression
Sprache des Titels:
Englisch
Original Buchtitel:
Computer Aided Systems Theory ? EUROCAST 2015. 15th International Conference, Las Palmas de Gran Canaria, Spain, February 8-13, 2015, Revised Selected Papers
Original Kurzfassung:
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. The result of the optimization is not a single solution anymore, but a whole Pareto-front describing the trade-off between accuracy and complexity.
In this contribution we study which complexity measures are most appropriately used in symbolic regression when performing multi-objective optimization with NSGA-II. Furthermore, we present a novel complexity measure that includes semantic information based on the function symbols occurring in the models and test its effects on several benchmark datasets. Results comparing multiple complexity measures are presented in terms of the achieved accuracy and model length to illustrate how the search direction of the algorithm is affected.