Genetic Learning
Genetic algorithms are the parts of evolutionary computing and are implemented as a computer simulation.
What is biological evolution?
Every organism has a set of rules which describes, how that organism is built, and encoded in the genes of an organism.
The chromosomes contain hundreds to thousands of genes.
Features of the organisms depend on the genes and they have several settings. For example: hair color gene may be brown or black.
The genes and their settings are referred to as a genotype of an organism.
When two organisms mate, they share their genes, so that the offspring gets half genes from one parent and half genes from the other parent. Hence, this process is called crossover.
It is possible that gene is mutated in the organism as a completely new feature.
The genetic algorithm follows the process of nature to solve the problems such as selection, crossover, mutation, and acceptance for evolution.
Following are the steps of genetic algorithm:
Step 1: Generate random population of ‘n’ chromosomes.
Step 2: Evaluate the fitness f(x) of each chromosome ‘x’ present in the population.
Step 3: Follow these steps to create a new population:
- a) Selection: From population, select two parents fitness according to their fitness.
- b) Crossover: Perform crossover using properties of crossover to form a new offspring.
- c) Mutation: Mutate the offspring at each position in chromosome.
- d) Accept: Place a new offspring in the new population.
Step 4: Replace: Use new generated population to run an algorithm.
Step 5: Test and Stop, if end condition is satisfied and return the best solution in the current population.
Step 6: Go to step 2.
