Abstract

Ant Colony Optimization (ACO) belongs to a class of biologically-motivated approaches to computing that includes such metaheuristics as artificial neural networks, evolutionary algorithms, and artificial immune systems, among others. Emulating to varying degrees the particular biological phenomena from which their inspiration is drawn, these alternative computational systems have succeeded in finding solutions to complex problems that had heretofore eluded more traditional techniques. Often, the resulting algorithm bears little resemblance to its biological progenitor, evolving instead into a mathematical abstraction of a singularly useful quality of the phenomenon. In such cases, these abstract computational models may be termed biological metaphors. Mindful that a fine line separates metaphor from distortion, this paper outlines an attempt to better understand the potential consequences an insufficient understanding of the underlying biological phenomenon may have on its transformation into mathematical metaphor. To that end, the author independently develops a rudimentary ACO, remaining as faithful as possible to the behavioral qualities of an ant colony. Subsequently, the performance of this new ACO is compared with that of a more established ACO in three categories: (1) the hybridization of evolutionary computing and ACO, (2) the efficacy of daemon actions, and (3) theoretical properties and convergence proofs. Ant Colony Optimization (ACO) belongs to a class of biologically-motivated approaches to computing that includes such metaheuristics as artificial neural networks, evolutionary algorithms, and artificial immune systems, among others. Emulating to varying degrees the particular biological phenomena from which their inspiration is drawn, these alternative computational systems have succeeded in finding solutions to complex problems that had heretofore eluded more traditional techniques. Often, the resulting algorithm bears little resemblance to its biological progenitor, evolving instead into a mathematical abstraction of a singularly useful quality of the phenomenon. In such cases, these abstract computational models may be termed biological metaphors. Mindful that a fine line separates metaphor from distortion, this paper outlines an attempt to better understand the potential consequences an insufficient understanding of the underlying biological phenomenon may have on its transformation into mathematical metaphor. To that end, the author independently develops a rudimentary ACO, remaining as faithful as possible to the behavioral qualities of an ant colony. Subsequently, the performance of this new ACO is compared with that of a more established ACO in three categories: (1) the hybridization of evolutionary computing and ACO, (2) the efficacy of daemon actions, and (3) theoretical properties and convergence proofs.

Notes

If this is your thesis or dissertation, and want to learn how to access it or for more information about readership statistics, contact us at STARS@ucf.edu

Graduation Date

2006

Semester

Summer

Degree

Master of Science (M.S.)

College

College of Sciences

Department

Mathematical Science

Format

application/pdf

Identifier

CFE0001192

URL

http://purl.fcla.edu/fcla/etd/CFE0001192

Language

English

Release Date

October 2018

Length of Campus-only Access

None

Access Status

Masters Thesis (Open Access)

Included in

Mathematics Commons

Share

COinS