- author = "Mark Kotanchek and Nathan Haut and Kelvin Kotanchek",
- title = "Representation and Reachability: Assumption Impact in Data Modeling",
- booktitle = "Genetic Programming Theory and Practice XXI",
- year = "2024",
- editor = "Stephan M. Winkler and Wolfgang Banzhaf and Ting Hu and Alexander Lalejini",
- series = "Genetic and Evolutionary Computation",
- pages = "1--25",
- address = "University of Michigan, USA",
- month = jun # " 6-8",
- publisher = "Springer",
- keywords = "genetic algorithms, genetic programming",
- isbn13 = "978-981-96-0076-2",
-
DOI = "
doi:10.1007/978-981-96-0077-9_1",
- abstract = "Data modeling implicitly makes assumptions. If those assumptions match the true system dynamics, the resulting model will be both insightful and predictive. If assumption alignment is not achieved, then we induce risk in using those models for prediction and action. Herein we argue that a good model consolidates energy (i.e., is parsimonious) and properly spans the interstitial regions between observed behavior points. From that viewpoint, we review and assess some of the common assumptions made in data modeling, in general, as well as in symbolic regression, in particular. We consider aspects of risk reduction, quality measures, functional building blocks, model representation, search operators, selection strategies, and trustability.",
-
notes = "Published in 2025 after the workshop
Evolved Analytics LLC, Rancho Santa Fe, CA, USA",
