The Effects of Population Size on the Performance of BEAGLE GPU-Based Genetic Programming Runs
Nathan Haut, Ilya Basin, Ruchika Gupta, Marzieh Kianinejad, Zachary Perrico + 2 more
TLDR
This paper investigates how population size impacts GPU-based Genetic Programming performance for symbolic regression, finding varied optimal strategies.
Key contributions
- Examined how GPU-enabled population sizes affect symbolic regression in Beagle GP.
- Showed constant population sizes benefit from narrow (1k) or broad (10M) searches depending on the problem.
- Explored stepped population strategies to balance search breadth and depth effectively.
Why it matters
This research provides crucial insights into optimizing GPU-based Genetic Programming by demonstrating the significant impact of population size. It highlights that optimal strategies vary by problem, offering guidance for more efficient and effective GP applications.
Original Abstract
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.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.