Statistical genetic programming for symbolic regression
Created by W.Langdon from
gpbibliography.bib Revision:1.8010
 @Article{journals/asc/HaeriEF17,

author = "Maryam Amir Haeri and Mohammad Mehdi Ebadzadeh and
Gianluigi Folino",

title = "Statistical genetic programming for symbolic
regression",

journal = "Applied Soft Computing",

year = "2017",

volume = "60",

pages = "447469",

month = nov,

keywords = "genetic algorithms, genetic programming, Symbolic
regression, Wellstructured subtree,
Semiwellstructured tree, Wellstructuredness measure,
Correlation coefficient",

bibdate = "20171122",

bibsource = "DBLP,
http://dblp.unitrier.de/db/journals/asc/asc60.html#HaeriEF17",

DOI = "doi:10.1016/j.asoc.2017.06.050",

abstract = "In this paper, a new genetic programming (GP)
algorithm for symbolic regression problems is proposed.
The algorithm, named statistical genetic programming
(SGP), uses statistical information (such as variance,
mean and correlation coefficient) to improve GP. To
this end, we define wellstructured trees as a tree
with the following property: nodes which are closer to
the root have a higher correlation with the target. It
is shown experimentally that on average, the trees with
structures closer to wellstructured trees are smaller
than other trees. SGP biases the search process to find
solutions whose structures are closer to a
wellstructured tree. For this purpose, it extends the
terminal set by some small wellstructured subtrees,
and starts the search process in a search space that is
limited to semiwellstructured trees (i.e., trees with
at least one wellstructured subtree). Moreover, SGP
incorporates new genetic operators, i.e.,
correlationbased mutation and correlationbased
crossover, which use the correlation between outputs of
each subtree and the targets, to improve the
functionality. Furthermore, we suggest a variancebased
editing operator which reduces the size of the trees.
SGP uses the new operators to explore the search space
in a way that it obtains more accurate and smaller
solutions in less time.
SGP is tested on several symbolic regression
benchmarks. The results show that it increases the
evolution rate, the accuracy of the solutions, and the
generalization ability, and decreases the rate of code
growth.",

notes = "Department of Computer Engineering and Information
Technology, Amirkabir University of Technology, Tehran,
Iran",
 }
Genetic Programming entries for
Maryam Amir Haeri
Mohammad Mehdi Ebadzadeh
Gianluigi Folino
Citations