Competition, Coevolution and the Game of Tag
Created by W.Langdon from
gp-bibliography.bib Revision:1.8178
- @InProceedings{Reynolds:1994:tag,
-
author = "Craig W. Reynolds",
-
title = "Competition, Coevolution and the Game of Tag",
-
booktitle = "Proceedings of the Fourth International Workshop on
the Synthesis and Simulation of Living Systems",
-
year = "1994",
-
editor = "Rodney A. Brooks and Pattie Maes",
-
pages = "59--69",
-
address = "MIT, Cambridge, MA, USA",
-
month = "6-8 " # jul,
-
publisher = "MIT Press",
-
keywords = "genetic algorithms, genetic programming",
-
URL = "http://www.cs.ucl.ac.uk/staff/W.Langdon/ftp/ftp.io.com/papers/cwrALifeIV.ps.Z",
-
size = "11 pages",
-
abstract = "Tag is a children's game based on symmetrical pursuit
and evasion. In these experiments, control programs for
mobile agents (simulated vehicles) are created through
artificial evolution, based on their skill at the game
of tag. Each controller is composed of distinct
behavior components for pursuit and evasion. A player's
fitness is based entirely on how well it performs in
competition with several other players chosen randomly
from the coevolving population of players. This
approach avoids the need for an expert player as a
fitness reference. In the beginning, the quality of
play is very low. This provides a fertile field for
slightly better strategies to exploit the weaknesses of
others. Through evolution, guided by competitive
fitness, increasingly better strategies emerge over
time.",
-
notes = "alife-4 Gives introduction to existing work on using
co-evolution in GPs and GAs. Uses Steady state GP with
tournament (7) selection. However 50% of deletions are
at random and 50% by inverse tournament. Some runs use
mutation. Some times uses Kinnear's Hoist crossover.
",
- }
Genetic Programming entries for
Craig W Reynolds
Citations