Chromosomes

In genetic algorithms, a chromosome is a set of parameters which define a proposed solution to the problem that the genetic algorithm is trying to solve. The chromosome is often represented as a simple string; although a wide variety of other data structures are also in use as chromosomes.

A genetic algorithm creates many chromosomes, either randomly
or by design, as an initial population. These chromosomes are each evaluated by the fitness function, which ranks them according to how good their solution is. The chromosomes which produced the best solutions, relatively speaking within the population, are allowed to breed, called crossover. The best chromosomes data is mixed, hopefully producing a better next generation.
The following are four basic forms of genetic strings which are commonly used.
Binary alphabet encoding
It is one of the most earliest and popular method, this may be an 8-bit binary coding of a real or integer number(s), or it may represent the position of switches and other devices. In this method, a mathematical function could be modeled and optimized. Or two CCD images could be matched for stereo vision, by how many pixels the objects have moved. It is not unknown to have strings that are 50, 80 or 100’s of elements long. For example: S=1001101101001010.
Gray coded encoding
Gray coded integers are similar to binary coding, but each consecutive number is represented by a difference of one bit. The advantage of gray coded integers is that in situations where the maxima between two numbers are been sought, for example gray code: 00000 and 10000 only differ in 1 bit. 7.
Integer or Real number encoding
It is a fairly obvious method where the values of each element in the strings are integer or real numbers. This will require the use of arrays to store all the values. These values usually represent the parameters to a particular function or problem. This method is not used a lot as many researcher prefer to use binary strings, as this allows for easier changes of the parameters themselves.
Symbolic coding
It is using of symbols to represent nodes on a tree. This method is used in node coding problems, such as in searching problems like the traveling salesman problem.
References:
Chromosome (genetic algorithm) - Wikipedia, the free encyclopedia, last modified 08:38, 2 February 2006, http://en.wikipedia.org/wiki/Chromosome_ (genetic_algorithm)
Book Reference: David Edward Goldberg (1989), “Genetic Algorithms in Search, Optimization and Machine Learning” Addison-Wesley.
Related Posts:

• ABSTRACT
• INTRODUCTION
• Genetic Algorithm
• Mechanism
• History
• Biological Background
• Parameters

Table of Contents
© 2006 Kumaravel & Project Team