Using an immune system model to explore mate selection in genetic algorithms. Page: 2 of 13
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to UNT Digital Library by the UNT Libraries Government Documents Department.
Extracted Text
The following text was automatically extracted from the image on this page using optical character recognition software:
Using an Immune System Model to Explore
Mate Selection in Genetic Algorithms
Chien-Feng Huang
Modeling. Algorithms, and Informatics Group (CCS-3).
Computer and Computational Sciences.
Los Alamos National Laboratory. MS B256,
Los Alamos. NM 87545. USA
cfhuangelanl.gov
Abstract. In the setting of multimodal function optimization, engineer-
ing and machine learning, identifying multiple peaks and maintaining
subpopulations of the search space are two central themes when Genetic
Algorithms (GAs) are employed. In this paper, an immune system model
is adopted to develop a framework for exploring the role of mate selection
in GAs with respect to these two issues. The experimental results re-
ported in the paper will shed more light into how mate selection schemes
compare to traditional selection schemes. In particular, we show that
dissimilar mating is beneficial in identifying multiple peaks, yet harmful
in maintaining subpopulations of the search space.
1 Introduction
In the setting of nmitimodal function optimization, engineering and machine
learning, there are two important issues when the GA is employed: (1) how fast
can the GA discover one or several peaks? And (2) can the GA maintain diverse
subpopulations in different parts of the search space?' In this paper. we intend to
use the mate-selection framework proposed in [7] and present the research work
for investigating these two themes. In [7], it was shown that mate selection plays
a crucial role in GA's search performance. In a nutshell, the dissimilarity-based
mate selection schemes facilitate locating a single, best-so-far solution at the
expense of generating lethal offspring; and the similarity-based mate selection
schemes enhance selection pressure toward highly-fit individuals such that the
GA's population converges rapidly to a certain region of a fitness landscape.
As such, for the first question. we would expect the dissimihrity-based mate
selection to improve the GA's search perfornimce with respect to that metric.
On the other hand, our empirical results so far have showed that simple GAs with
the mate selection schemes are all subject to convergence (i.e.. the simple GAs
The first issue was briefly discussed in [7]. For the second issue, there are some
practical problems where maintaining subpopulations are critical. An example is the
application of genetic approach to decentralized PI controller tuning for multivariable
processes in [12].
Upcoming Pages
Here’s what’s next.
Search Inside
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Huang, C. F. (Chien-Feng). Using an immune system model to explore mate selection in genetic algorithms., article, January 1, 2003; United States. (https://digital.library.unt.edu/ark:/67531/metadc933202/m1/2/: accessed April 25, 2024), University of North Texas Libraries, UNT Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.