Multi-population adapative inflationary differential evolution

Research output: Contribution to conferencePaper

Abstract

In this paper, a multi-population version of Adaptive Inflationary Differential Evolution, which automatically adapts the crossover probability and the differential weight of the Differential Evolution, is presented. The multi-population algorithm exploits the use of different populations, and the local minima found by each population, to assess the distance between minima; a probabilistic kernel based approach is then used to automatically adapt the dimension of a bubble in which the population is re-initialized after converging to a local minimum. The algorithm is tested on two real case functions and on two difficult academic functions.

Conference

ConferenceBio-inspired Optimization Methods and their Applications, BIOMA 14
CountrySlovenia
CityLjubljana
Period13/09/1413/09/14

Keywords

  • adaptive algorithms
  • differential evolution
  • global optimization

Cite this

Di Carlo, M., Vasile, M., & Minisci, E. (2014). Multi-population adapative inflationary differential evolution. 41-54. Paper presented at Bio-inspired Optimization Methods and their Applications, BIOMA 14, Ljubljana, Slovenia.
Di Carlo, Marilena ; Vasile, Massimiliano ; Minisci, Edmondo. / Multi-population adapative inflationary differential evolution. Paper presented at Bio-inspired Optimization Methods and their Applications, BIOMA 14, Ljubljana, Slovenia.
@conference{fef0a381e18e43f4ae50db261f0f755d,
title = "Multi-population adapative inflationary differential evolution",
abstract = "In this paper, a multi-population version of Adaptive Inflationary Differential Evolution, which automatically adapts the crossover probability and the differential weight of the Differential Evolution, is presented. The multi-population algorithm exploits the use of different populations, and the local minima found by each population, to assess the distance between minima; a probabilistic kernel based approach is then used to automatically adapt the dimension of a bubble in which the population is re-initialized after converging to a local minimum. The algorithm is tested on two real case functions and on two difficult academic functions.",
keywords = "adaptive algorithms, differential evolution, global optimization",
author = "{Di Carlo}, Marilena and Massimiliano Vasile and Edmondo Minisci",
year = "2014",
month = "9",
day = "13",
language = "English",
pages = "41--54",
note = "Bio-inspired Optimization Methods and their Applications, BIOMA 14 ; Conference date: 13-09-2014 Through 13-09-2014",

}

Di Carlo, M, Vasile, M & Minisci, E 2014, 'Multi-population adapative inflationary differential evolution' Paper presented at Bio-inspired Optimization Methods and their Applications, BIOMA 14, Ljubljana, Slovenia, 13/09/14 - 13/09/14, pp. 41-54.

Multi-population adapative inflationary differential evolution. / Di Carlo, Marilena; Vasile, Massimiliano; Minisci, Edmondo.

2014. 41-54 Paper presented at Bio-inspired Optimization Methods and their Applications, BIOMA 14, Ljubljana, Slovenia.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Multi-population adapative inflationary differential evolution

AU - Di Carlo,Marilena

AU - Vasile,Massimiliano

AU - Minisci,Edmondo

PY - 2014/9/13

Y1 - 2014/9/13

N2 - In this paper, a multi-population version of Adaptive Inflationary Differential Evolution, which automatically adapts the crossover probability and the differential weight of the Differential Evolution, is presented. The multi-population algorithm exploits the use of different populations, and the local minima found by each population, to assess the distance between minima; a probabilistic kernel based approach is then used to automatically adapt the dimension of a bubble in which the population is re-initialized after converging to a local minimum. The algorithm is tested on two real case functions and on two difficult academic functions.

AB - In this paper, a multi-population version of Adaptive Inflationary Differential Evolution, which automatically adapts the crossover probability and the differential weight of the Differential Evolution, is presented. The multi-population algorithm exploits the use of different populations, and the local minima found by each population, to assess the distance between minima; a probabilistic kernel based approach is then used to automatically adapt the dimension of a bubble in which the population is re-initialized after converging to a local minimum. The algorithm is tested on two real case functions and on two difficult academic functions.

KW - adaptive algorithms

KW - differential evolution

KW - global optimization

UR - http://bioma.ijs.si/conference/2014/

M3 - Paper

SP - 41

EP - 54

ER -

Di Carlo M, Vasile M, Minisci E. Multi-population adapative inflationary differential evolution. 2014. Paper presented at Bio-inspired Optimization Methods and their Applications, BIOMA 14, Ljubljana, Slovenia.