Application of Genetic Programming to Unsolved Mathematical Problems II

In this research, the authors developed a new Swift programming library of symbolic regression based on genetic programming. Symbolic regression is a field of artificial intelligence where AI looks for formulas that describe the given data. The data used for the research is winning positions of comb...

Full description

Saved in:
Bibliographic Details
Main Authors: Tanemura, Keito, Sasaki, Yuji, Takahashi, Shoei, Tokuni, Yuki, Manabe, Hikaru, Miyadera, Ryohei
Format: Conference Proceeding
Language:English
Subjects:
combinatorial games
Electric potential
Games
Genetic programming
Integer programming
Libraries
mixed integer programming
Programming
Search problems
symbolic regression
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this research, the authors developed a new Swift programming library of symbolic regression based on genetic programming. Symbolic regression is a field of artificial intelligence where AI looks for formulas that describe the given data. The data used for the research is winning positions of combinatorial games. Compared to the library presented at the last GCCE conference, this library gets two new features to select the fittest formulae.The first is to find a minimum number of formulae that describe the given data.The second is to separate the data into smaller subsets, and find formulae to describe each subset.With these two new features, this new Swift programming library of symbolic regression can be a powerful tool in the research of mathematics and science.
ISSN:2693-0854
DOI:10.1109/GCCE59613.2023.10315661