Investigating the parameter space of evolutionary algorithms

被引：54

作者：

Sipper, Moshe ^{[1
,2
]}

Fu, Weixuan ^{[1
]}

Ahuja, Karuna ^{[1
]}

Moore, Jason H. ^{[1
]}

机构：

[1] Univ Penn, Inst Biomed Informat, Philadelphia, PA 19104 USA

[2] Ben Gurion Univ Negev, Dept Comp Sci, IL-8410501 Beer Sheva, Israel

来源：

BIODATA MINING | 2018年 / 11卷

基金：

美国国家卫生研究院;

关键词：

Evolutionary algorithms; Genetic programming; Meta-genetic algorithm; Parameter tuning; Hyper-parameter; OPTIMIZATION; SEARCH;

D O I：

10.1186/s13040-018-0164-x

中图分类号：

Q [生物科学];

学科分类号：

07 ; 0710 ; 09 ;

摘要：

Evolutionary computation (EC) has been widely applied to biological and biomedical data. The practice of EC involves the tuning of many parameters, such as population size, generation count, selection size, and crossover and mutation rates. Through an extensive series of experiments over multiple evolutionary algorithm implementations and 25 problems we show that parameter space tends to be rife with viable parameters, at least for the problems studied herein. We discuss the implications of this finding in practice for the researcher employing EC.

引用

页数：14

共 33 条

[1]

[Anonymous], EVOSTAR 2017

[2]

[Anonymous], EVOLUTION RUNS FASTE

[3]

[Anonymous], GENETIC PROGRA UNPUB

[4]

[Anonymous], 2005, PROBLEM DEFINITIONS

[5] Parameter tuning or default values? An empirical investigation in search-based software engineering [J].