A decomposition method for symbolic regression problems

Created by W.Langdon from gp-bibliography.bib Revision:1.7630

@Article{MOUSAVIASTARABADI:2018:ASC,

• author = "Samaneh Sadat {Mousavi Astarabadi} and Mohammad Mehdi Ebadzadeh",
• title = "A decomposition method for symbolic regression problems",
• journal = "Applied Soft Computing",
• volume = "62",
• pages = "514--523",
• year = "2018",
• keywords = "genetic algorithms, genetic programming, Symbolic regression, Performance estimation, Decomposition, Optimization",
• ISSN = "1568-4946",
• DOI = "doi:10.1016/j.asoc.2017.10.041",
• URL = "http://www.sciencedirect.com/science/article/pii/S1568494617306555",
• abstract = "The purpose of this paper is to improve the efficiency of Genetic Programming (GP) by decomposing a regression problem into several subproblems. An optimization problem is defined to find subproblems of the original problem for which the performance of GP is better than for the original problem. In order to evaluate the proposed decomposition method, the subproblems of several benchmark problems are found by solving the optimization problem. Then, a 2-layer GP system is used to find subproblems' solutions in the first layer and the solution of the original problem in the second layer. The results of this 2-layer GP system show that the proposed decomposition method does not generate trivial subproblems. It generates subproblems that improve the efficiency of GP against when subproblems are not used",

}

Genetic Programming entries for Samaneh Sadat Mousavi Astarabadi Mohammad Mehdi Ebadzadeh

Citations