Artificial Intelligent Systems and Machine Learning

Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Analyzing the Effect of Sigma Scaling in Genetic Algorithms


Affiliations
1 Department of Computer Engineering, University Institute of Engineering and Technology, Kurukshetra University, Kurukshetra, India
2 Department of Mechanical Engineering, University Institute of Engineering and Technology, Kurukshetra University, Kurukshetra, India
     

   Subscribe/Renew Journal


The Genetic Algorithms (GA) paradigm is being used increasingly in search and optimization problems. A genetic algorithm is an algorithm which varies a set of parameters and evaluates the quality or "fitness" of the results of a computation as the parameters are changed or "evolved". A study of the effect of Sigma scaling on the performance of Genetic Algorithm (GA) is reported here. This paper shows how the Sigma Scaling affects the performance of Genetic Algorithm by calculating the average and maximum fitness values. It is beneficial to use sigma scaling with genetic algorithms for three reasons. It prevents the Genetic Algorithm from premature convergence, maintains diversity in the population and always improves the performance of GA. Premature convergence is one of the major problems usually associated with the use of GA. This problem is tightly related with the loss of genetic diversity of the GA's population, being the cause of a decrease on the quality of the solutions found.

Keywords

Crossover, Fitness Function, Genetic Algorithm, Mutation, Sigma Scaling, Selection.
User
Subscription Login to verify subscription
Notifications
Font Size

Abstract Views: 330

PDF Views: 3




  • Analyzing the Effect of Sigma Scaling in Genetic Algorithms

Abstract Views: 330  |  PDF Views: 3

Authors

Chander Diwaker
Department of Computer Engineering, University Institute of Engineering and Technology, Kurukshetra University, Kurukshetra, India
Upender Dhull
Department of Mechanical Engineering, University Institute of Engineering and Technology, Kurukshetra University, Kurukshetra, India
Geeta Jangra
Department of Computer Engineering, University Institute of Engineering and Technology, Kurukshetra University, Kurukshetra, India

Abstract


The Genetic Algorithms (GA) paradigm is being used increasingly in search and optimization problems. A genetic algorithm is an algorithm which varies a set of parameters and evaluates the quality or "fitness" of the results of a computation as the parameters are changed or "evolved". A study of the effect of Sigma scaling on the performance of Genetic Algorithm (GA) is reported here. This paper shows how the Sigma Scaling affects the performance of Genetic Algorithm by calculating the average and maximum fitness values. It is beneficial to use sigma scaling with genetic algorithms for three reasons. It prevents the Genetic Algorithm from premature convergence, maintains diversity in the population and always improves the performance of GA. Premature convergence is one of the major problems usually associated with the use of GA. This problem is tightly related with the loss of genetic diversity of the GA's population, being the cause of a decrease on the quality of the solutions found.

Keywords


Crossover, Fitness Function, Genetic Algorithm, Mutation, Sigma Scaling, Selection.