- author = "Illya Bakurov and Aidan Murphy and Wolfgang Banzhaf",
- title = "Analyzing Fitness Aggregation Strategies for Symbolic Regression Problem-Solving",
- booktitle = "Genetic Programming Theory and Practice XXII",
- year = "2025",
- editor = "Bogdan Burlacu and Fabricio {Olivetti de Franca} and Alexander Lalejini and Stephen Kelly and Wolfgang Banzhaf",
- series = "Genetic and Evolutionary Computation",
- pages = "23--44",
- address = "Michigan State University, USA",
- month = jun # " 5-7",
- publisher = "Springer",
- keywords = "genetic algorithms, genetic programming",
- isbn13 = "978-981-95-6397-5",
-
DOI = "
10.1007/978-981-95-6398-2_2",
-
code_url = "
https://gitlab.com/ibakurov/general-purpose-optimization-library",
- abstract = "... Using the coefficient of determination () as the fitness function relieves GP from evolving the scale and translation factors and allows GP to focus on identifying the correct shape/functional form of the target function. While demonstrates promising generalization capabilities and eliminates the need for per-individual linear scaling, it remains limited to detecting linear relationships, it is sensitive to noise, and can be zero even when two variables are statistically dependent. we extend the exploration of correlation-based fitness assessment beyond by evaluating alternative measures of statistical dependence (such as Spearman Rank Correlation Coefficient and Distance Correlation), and similarity (such as Cosine Similarity). Furthermore, we explore combining these functions through multi-objectivePareto optimization. Our experimental results on both synthetic and real-world datasets demonstrate that these alternatives can improve generalization ability of the SR models by capturing meaningful, non-linear structural similarities overlooked by alone",
-
notes = "lightning talk.
published in 2026 after the workshop",
