Detail výsledku
Graph-based Genetic Programming
Dal Piccol Sotto Léo Françoso
Vašíček Zdeněk, doc. Ing., Ph.D., UPSY (FIT)
Although the classical way to represent programs in Genetic Programming (GP) is by means of an expression tree, different GP variants with alternative representations have been proposed throughout the years. One such representation is the Directed Acyclic Graph (DAG), adopted by methods like Cartesian Genetic Programming (CGP), Linear Genetic Programming (LGP), Parallel Distributed Genetic Programming (PDGP), and, more recently, Evolving Graphs by Graph Programming (EGGP). The aim of this tutorial is to consider this methods from a unified perspective as graph-based GP, present their historical background, representation features, operators, applications, and available implementations.
Genetic Programming, Cartesian Genetic Programming, Linear Genetic Programming, Parallel Distributed Genetic Programming
@inproceedings{BUT180545,
author="Roman {Kalkreuth} and Léo Françoso {Dal Piccol Sotto} and Zdeněk {Vašíček}",
title="Graph-based Genetic Programming",
booktitle="GECCO 2022 Companion - Proceedings of the 2022 Genetic and Evolutionary Computation Conference",
year="2022",
pages="958--982",
publisher="Association for Computing Machinery",
address="Boston",
doi="10.1145/3520304.3533657",
isbn="978-1-4503-9268-6"
}