The MAX Problem for Genetic Programming -
Highlighting an Adverse Interaction between the
Crossover Operator and a Restriction on Tree Depth
C. Gathercole, and P. Ross. Department of Artificial Intelligence, University of
Edinburgh, 80 South Bridge, Edinburgh, EH1 1HN, UK, (1995)
Abstract
The Crossover operator is common to most
implementations of Genetic Programming (GP). Another,
usually unavoidable, factor is some form of restriction
on the size of trees in the GP population. This paper
concentrates on the interaction between the Crossover
operator and a restriction on tree depth demonstrated
by the MAX problem, which involves returning the
largest possible value for given function and terminal
sets. Some characteristics and inadequacies of
Crossover in `normal' use are...
p.s. On a related theme, and only blowing my own
trumpet a little bit, I have recently written a paper
Gathercole (soon to be submitted to GP96) which looks
at an unfortunate interaction in GP between the
Crossover operator and restrictions on tree size.
Its more or less finished
Published as Gathercole:1996:aicrtd
%0 Report
%1 Gathercole
%A Gathercole, Chris
%A Ross, Peter
%C 80 South Bridge, Edinburgh, EH1 1HN, UK
%D 1995
%K algorithms, genetic programming
%T The MAX Problem for Genetic Programming -
Highlighting an Adverse Interaction between the
Crossover Operator and a Restriction on Tree Depth
%U http://citeseer.ist.psu.edu/gathercole95max.html
%X The Crossover operator is common to most
implementations of Genetic Programming (GP). Another,
usually unavoidable, factor is some form of restriction
on the size of trees in the GP population. This paper
concentrates on the interaction between the Crossover
operator and a restriction on tree depth demonstrated
by the MAX problem, which involves returning the
largest possible value for given function and terminal
sets. Some characteristics and inadequacies of
Crossover in `normal' use are...
@techreport{Gathercole,
abstract = {The Crossover operator is common to most
implementations of Genetic Programming (GP). Another,
usually unavoidable, factor is some form of restriction
on the size of trees in the GP population. This paper
concentrates on the interaction between the Crossover
operator and a restriction on tree depth demonstrated
by the MAX problem, which involves returning the
largest possible value for given function and terminal
sets. Some characteristics and inadequacies of
Crossover in `normal' use are...},
added-at = {2008-06-19T17:35:00.000+0200},
address = {80 South Bridge, Edinburgh, EH1 1HN, UK
},
author = {Gathercole, Chris and Ross, Peter},
biburl = {https://www.bibsonomy.org/bibtex/2153f2be0a2da2614f40100eba817f6f2/brazovayeye},
broken = {ftp://ftp.dai.ed.ac.uk/pub/user/chrisg/max-problem-in-GP.for_submission_to_gp-96.ps.gz},
institution = {Department of Artificial Intelligence, University of
Edinburgh},
interhash = {5480c59219bbcff5a57c7e6170b67f5e},
intrahash = {153f2be0a2da2614f40100eba817f6f2},
keywords = {algorithms, genetic programming},
notes = {p.s. On a related theme, and only blowing my own
trumpet a little bit, I have recently written a paper
[Gathercole] (soon to be submitted to GP96) which looks
at an unfortunate interaction in GP between the
Crossover operator and restrictions on tree size.
Its more or less finished
Published as \cite{Gathercole:1996:aicrtd}},
size = {10 pages},
timestamp = {2008-06-19T17:40:08.000+0200},
title = {The {MAX} Problem for Genetic Programming -
Highlighting an Adverse Interaction between the
Crossover Operator and a Restriction on Tree Depth},
url = {http://citeseer.ist.psu.edu/gathercole95max.html},
year = 1995
}