Saved in:
Bibliographic Details
Main Authors: Haut, Nathan, Basin, Ilya, Gupta, Ruchika, Kianinejad, Marzieh, Perrico, Zachary, Smith, Elijah, Banzhaf, Wolfgang
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.24968
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917442382462976
author Haut, Nathan
Basin, Ilya
Gupta, Ruchika
Kianinejad, Marzieh
Perrico, Zachary
Smith, Elijah
Banzhaf, Wolfgang
author_facet Haut, Nathan
Basin, Ilya
Gupta, Ruchika
Kianinejad, Marzieh
Perrico, Zachary
Smith, Elijah
Banzhaf, Wolfgang
contents The Beagle framework, through GPU-based Genetic Programming, enables population dynamics previously unattainable (within practical time frames) by CPU-constrained Genetic Programming systems. This work explores how GPU-enabled population sizes impact the success of training for symbolic regression problems. Specifically, when using constant population sizes, we see benefits of using very narrow and deep searches (as narrow as 1000 individuals) for some problems, while other problems benefit from very broad and shallow searches (as broad as 10 million individuals). We also explore stepped population sizes that start with large populations and drop to small populations to balance the breadth and depth of search.
format Preprint
id arxiv_https___arxiv_org_abs_2604_24968
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle The Effects of Population Size on the Performance of BEAGLE GPU-Based Genetic Programming Runs
Haut, Nathan
Basin, Ilya
Gupta, Ruchika
Kianinejad, Marzieh
Perrico, Zachary
Smith, Elijah
Banzhaf, Wolfgang
Neural and Evolutionary Computing
The Beagle framework, through GPU-based Genetic Programming, enables population dynamics previously unattainable (within practical time frames) by CPU-constrained Genetic Programming systems. This work explores how GPU-enabled population sizes impact the success of training for symbolic regression problems. Specifically, when using constant population sizes, we see benefits of using very narrow and deep searches (as narrow as 1000 individuals) for some problems, while other problems benefit from very broad and shallow searches (as broad as 10 million individuals). We also explore stepped population sizes that start with large populations and drop to small populations to balance the breadth and depth of search.
title The Effects of Population Size on the Performance of BEAGLE GPU-Based Genetic Programming Runs
topic Neural and Evolutionary Computing
url https://arxiv.org/abs/2604.24968