Genetic Algorithms: Stopping Times and a New Model.

Date of Submission

December 2000

Date of Award

Winter 12-12-2001

Institute Name (Publisher)

Indian Statistical Institute

Document Type

Master's Dissertation

Degree Name

Master of Technology

Subject Name

Computer Science

Department

Machine Intelligence Unit (MIU-Kolkata)

Supervisor

Murthy, C. A. (MIU-Kolkata; ISI)

Abstract (Summary of the Work)

Mankind has been trying to solve optimization problems in different spheres of life. There have been several ways in which those problems have been attempted previously. Some of these are:• Calculus based techniques: use indirect methods where a set of non-linear equa- tions is solved or direct methods which are tools like hill climbing.• Enumerative techniques: conceptually very simple, involving evaluation of ob- jective function at every point of the search space.• Random techniques: search space is inVestigated at random.Genetic Algorithms(GAs) are stochastic search methods belonging to the last category, random techniques. They perforin a multi-dimensional scarch in very large, complex and multimodal scarch spaces in providing an optimal solution for evalua- tion(fitness) function of an optimization problem. The idea of GAs originated from the principles of natural genetic systems, particularly from the theory of evolution. GAs are empirically found to provide global near optimai solutions to various complex optimization problems when the search space is discrete.Genetic Algoritbms came into existence since researchers wanted to simulate the natural genetic systems on computer. Applying GAs to solve optimization problems became a natural use of GAs since the human genetic system is supposed to be the best in the world and it evolved over various generations. While solving an opti- mization problem using GAs, esch potential solution is encoded as a string(called chromosome) of finite length (say, L) over a finite alphabet A. Each string or chro- mosome is considered as an individual. A collection of M(M is finite) such individuals is called a population. Usually, three naturally occurring phenomena are incorporated in GAs. The three phenomena are• Reproduction/selection• Crossover• Mutation.The above biologically inspired phenomena are applied on the chromosomes to yield potentially better chromosomes. In each generation(mathematicaliy iteration) a new population of the same size is generated from the current population using the above phenomena and this new population is then used to generate another popu- lation. Note that the number of possible populations is finite as M, L,and A are finite.Some of the application areas of GAs include operations research, pattern classi- fication and feature selection, image processing|10] and scene recognition, rule gener- ation and classifier systems(9], neural network design{8], acheduling problems, VLSI design, path planning(7] and the travelling salesman problem[6), graph coloring|5), numerical optimization.There are several problems in applying genetic algorithms for real life applications. The problem of finding the stopping time for GAs and obtaining a new model in this regard are attempted here.

Comments

ProQuest Collection ID: http://gateway.proquest.com/openurl?url_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&res_dat=xri:pqm&rft_dat=xri:pqdiss:28843305

Control Number

ISI-DISS-2000-77

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

DOI

http://dspace.isical.ac.in:8080/jspui/handle/10263/6249

This document is currently not available here.

Share

COinS