Large language models help programs to evolve

Usingthis representation, a genetic-programming system 'mutates' a program by randomly changing one node in the tree to a different value (Fig. la). Instead of replacing random parts of a syntax tree, an LLM can generate a variation of a program written in a standard programming language,...

Full description

Saved in:
Bibliographic Details
Published inNature (London) Vol. 625; no. 7995; pp. 452 - 453
Main Author Mouret, Jean-Baptiste
Format Journal Article
LanguageEnglish
Published London Nature Publishing Group 18.01.2024
Subjects
Artificial intelligence
Computer science
Computers
Genetic algorithms
Large language models
Mathematical models
Programmers
Programming languages
Python
Syntax
Online AccessGet full text
ISSN0028-0836
1476-4687
DOI10.1038/d41586-023-03998-0

Cover

More Information
Summary:Usingthis representation, a genetic-programming system 'mutates' a program by randomly changing one node in the tree to a different value (Fig. la). Instead of replacing random parts of a syntax tree, an LLM can generate a variation of a program written in a standard programming language, such as Python. To do so, a simple, but powerful, approach is to select two programs, concatenate them, and ask the LLM to complete the program using the concatenated pair as a prompt - resulting in the generation of a third program (Fig. 1b). Romera-Paredes et al. used this fresh approach to genetic programming to find ways of solving mathematical problems in optimization and geometry that were better than the best attempts of human programmers.
Bibliography:SourceType-Scholarly Journals-1
ObjectType-News-1
content type line 14
ISSN:0028-0836
1476-4687
DOI:10.1038/d41586-023-03998-0