Introduction to Genetic Algorithms: Theory and Applications

This section first goes through the process of representing genes in the GA algorithm. Then, the fitness function is introduced which evaluates of the chromosomes in the GA algorithm. The learning outcomes are as follows:  

• Understanding the reason of gene representation in GA 
• Understanding the reason of using a fitness function to evaluate chromosomes 
• Understanding the process of representing genes 
• Understanding the process of evaluating choromosmes using a fitness function

This coding video shows the steps of representing genes and evaluating them using a fitness function. The learning outcomes are given:  

• Understanding the process of coding genes in the Genetic Algorithm 
• Understanding the steps of preparing a fitness function
• Understanding the process of creating an initial population of chromosomes with random genes 

This lecture is about the most popular selection mechanism in the Genetic Algorithm called roulette wheel. The mathematical model behind a roulette wheel is discussed in details and used to create a virtual roulette wheel. This roulette wheel is then employed to select chromosomes proportional to their fitness value. The learning objectives are:

• Understanding the mechanism of the roulette wheel operator in GA
  
• Underestaring the selection mechanism of GA using roulette wheel 
• Understanding the reason of using a selection operator in GA

This lecture takes you through an example of selection process in the GA algorithm. A set of chromosomes will be used as an example. Therefore, you will practice how to implement the roulette wheel and selection mechanism of GA. 

• Understanding the process of implementing a roulette wheel in Matlab 
• Understanding the process of selecting chromosomes using a roulette wheel in Matlab 

This coding video shows you the steps of developing a function to select two parents from a given population (generation). This function is tested and analyzed to be used later on when developing the GA algorithm.  The learning objectives are: 

• Understanding the process of encapsulating the roulette wheel code in a function 
• Demonstrating the ability to use the selection function to select two parents. 
• Demonstrating the ability to test, debug, and run the selection function in Matlab. 

The Roulette wheel does not work when fitness values are negative. This lectures covers a method called scaling to fix this issue.

In the last module of this course, we developed a selection function to select two parents. This lecture shows you different methods of combining the parents and producing children. The main learning outcomes are: 

• Understand the single-point crossover 
• Understanding the double-point crossover 
• Understanding the reason of using the probability of crossover (Pc) to accept children 

This lecture provides the step-by-step of coding single- and double-point crossover (recombination) techniques in the Matlab programming language. The selection mechanism is first employed to create two parents. Then, the crossover is done to product two children. You will also see how to accept and transfer a newly created child using the probability of crossover. The learning outcomes are: 

• Implementing the single-point crossover in Matlab
• Implementing the double-point crossover in Matlab
• Understanding the use and implementation of the probability crossover 

In the lecture, we will be developing a function to perform different types of crossover between two given parents and return wither children or parents using the probability of crossover. The learning outcomes area: 

• Understanding the steps of designing a function to recombine two parents. 
• Understanding the process of making a multi-purpose crossover function.
• Understanding the use of a crossover function to perform crossover in the GA algorithm. 

This lecture covers the mutation operator in GA that causes random changes in the genes of chromosomes. The idea behind the mutation is first discussed. Then, the process of applying this operator to genes using the probability of mutation (Pc) is given. The learning outcomes of this lecture are: 

• Understanding the necessity of mutation to increase the diversity of a population of individuals 
• Understanding the process of using the probability of mutation to mutate genes in a chromosome 

This lecture is a tutorial on coding a function in Matlab for mutating the genes of a given chromosome. The function is developed with two inputs: chromosome and probability of mutation. Depending on the Pm and a generated random number, the genes might face mutation or not. The learning outcomes are: 

• Understanding the process designing a function to mutation the genes 
  
• Underesating the use of Pm to decide whether a gene should be mutated or not

This lecture covers the rational and idea behind elitism in the Genetic Algorithm. A parameter called Elitism ration (Er) is also presented, which shows the portion of the best individuals that should be kept as elited and get transferred to the next population (generation) unchanged. The learning outcomes area as follows: 

• Understanding the rationals behind the elitism operator 
• Understanding the use of Er in defining the number of elites 
• Understanding the process of elitism and its impact on a subsequent population 
  

This lecture shows you an example of an elitism operator in the Genetic Algorithm. The learning outcomes are: 

• Understanding the process of saving the best chromosomes as elites in the Genetic Algorithm 
• Understanding the use and implementation of Elitism ration (Er) 

In this lecture, a Matlab function has been developed to perform elitism on a given population using the elitism ration. The learning outcomes are: 

• To understand the mechanism of a Matlab function to save the best chromoems 
• To understand the process of applying the eilitism function to a given population

This lecture improves the code written for the elitism in the last video. The main objective is to make sure that the elites never get lost or damaged during the process of crossover or mutation. 

This video develops the first drafts of the Genetic Algorithm using the function that we designed before. 

This lecture provides the steps of designing a function for the Genetic Algorithm that facilitates the process of applying this algorithm to optimization problems.