Geometric semantic GP with linear scaling: Darwinian versus Lamarckian evolution
Created by W.Langdon from
gp-bibliography.bib Revision:1.7970
- @Article{Nadizar:2024:GPEM,
-
author = "Giorgia Nadizar and Berfin Sakallioglu and
Fraser Garrow and Sara Silva and Leonardo Vanneschi",
-
title = "Geometric semantic {GP} with linear scaling:
{Darwinian} versus {Lamarckian} evolution",
-
journal = "Genetic Programming and Evolvable Machines",
-
year = "2024",
-
volume = "25",
-
pages = "Article no: 17",
-
note = "Online first",
-
keywords = "genetic algorithms, genetic programming, Symbolic
regression, Geometric semantic genetic programming,
Linear scaling, Lamarckian evolution",
-
ISSN = "1389-2576",
-
URL = "https://rdcu.be/dJI7W",
-
DOI = "doi:10.1007/s10710-024-09488-0",
-
abstract = "Geometric Semantic Genetic Programming (GSGP) has
shown notable success in symbolic regression with the
introduction of Linear Scaling (LS). This achievement
stems from the synergy of the geometric semantic
genetic operators of GSGP with the scaling of the
individuals for computing their fitness, which favours
programs with a promising behaviour. However, the
initial combination of GSGP and LS (GSGP-LS) underused
the potential of LS, scaling individuals only for
fitness evaluation, neglecting to incorporate
improvements into their genetic material. In this paper
we propose an advancement, GSGP with Lamarckian LS
(GSGP-LLS), wherein we update the individuals in the
population with their scaling coefficients in a
Lamarckian fashion, i.e., by inheritance of acquired
traits. We assess GSGP-LS and GSGP-LLS against standard
GSGP for the task of symbolic regression on five
hand-tailored benchmarks and six real-life problems. On
the former ones, GSGP-LS and GSGP-LLS both consistently
improve GSGP, though with no clear global superiority
between them. On the real-world problems, instead,
GSGP-LLS steadily outperforms GSGP-LS, achieving faster
convergence and superior final performance. Notably,
even in cases where LS induces overfitting on
challenging problems, GSGP-LLS surpasses GSGP-LS, due
to its slower and more localised optimisation steps.",
-
notes = "Department of Mathematics, Informatics, and
Geosciences, University of Trieste, Trieste, Italy",
- }
Genetic Programming entries for
Giorgia Nadizar
Berfin Sakallioglu
Fraser Garrow
Sara Silva
Leonardo Vanneschi
Citations