Genetic algorithms: Difference between revisions

From Computer Science Wiki
No edit summary
 
(11 intermediate revisions by 2 users not shown)
Line 2: Line 2:


The genetic algorithm is a method for solving both constrained and unconstrained optimization problems that is based on natural selection, the process that drives biological evolution. The genetic algorithm repeatedly modifies a population of individual solutions. At each step, the genetic algorithm selects individuals at random from the current population to be parents and uses them to produce the children for the next generation. Over successive generations, the population "evolves" toward an optimal solution. You can apply the genetic algorithm to solve a variety of optimization problems that are not well suited for standard optimization algorithms, including problems in which the objective function is discontinuous, nondifferentiable, stochastic, or highly nonlinear. The genetic algorithm can address problems of mixed integer programming, where some components are restricted to be integer-valued.<ref>https://www.mathworks.com/help/gads/what-is-the-genetic-algorithm.html</ref>
The genetic algorithm is a method for solving both constrained and unconstrained optimization problems that is based on natural selection, the process that drives biological evolution. The genetic algorithm repeatedly modifies a population of individual solutions. At each step, the genetic algorithm selects individuals at random from the current population to be parents and uses them to produce the children for the next generation. Over successive generations, the population "evolves" toward an optimal solution. You can apply the genetic algorithm to solve a variety of optimization problems that are not well suited for standard optimization algorithms, including problems in which the objective function is discontinuous, nondifferentiable, stochastic, or highly nonlinear. The genetic algorithm can address problems of mixed integer programming, where some components are restricted to be integer-valued.<ref>https://www.mathworks.com/help/gads/what-is-the-genetic-algorithm.html</ref>
<br />


[https://www.analyticsvidhya.com/blog/2017/07/introduction-to-genetic-algorithm/ This article is a good place to get started]
== Two videos to get you started ==
<html>
<iframe width="560" height="315" src="https://www.youtube.com/embed/05rEefXlmhI" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
</html>
<br />
<html>
<iframe width="560" height="315" src="https://www.youtube.com/embed/1i8muvzZkPw" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
</html>


=== Initial population ===
== Use of genetic algorithms ==  
=== Fitness function ===
* [https://en.wikipedia.org/wiki/List_of_genetic_algorithm_applications Click here for a fairly good list]
=== Selection ===
* [https://stackoverflow.com/questions/1538235/what-are-good-examples-of-genetic-algorithms-genetic-programming-solutions Click here for several good examples (scroll down)]
Selection rules select the individuals, called parents, that contribute to the population at the next generation.
* [https://www.brainz.org/15-real-world-applications-genetic-algorithms/ here are some short examples of GA]


=== Crossover ===
== The basic pattern of genetic algorithms ==  
Crossover rules combine two parents to form children for the next generation.
# A random set of solutions would be generated on the sample documents
# And tested against the human labelling
# Best fit solutions retained
# New generation created by mutating/crossing
# Algorithm repeated
# Until a good fit obtained


=== Mutation ===
== Terms associated with genetic algorithms ==
Mutation rules apply random changes to individual parents to form children.
 
* [[Brute force approach]]
* [[Combinatorial optimization]]
* [[Computational intractability]]
* [[Convergence]]
* [[Crossover / crossover operator]]
* [[Elitism]]
* [[Exploration vs exploitation]]
* [[Fitness / fitness function / fitness landscape]]
* [[Heuristics]]
* [[Hill climbing]]
* [[Initialization parameters]]
* [[Local extrema]]
* [[Mating pool]]
* [[Mutation / mutation rate]]
* [[Novelty search]]
* [[Offspring]]
* [[Optimization]]
* [[Population]]
* [[Premature convergence]]
* [[Problem space]]
* [[Ranking]]
* [[Roulette wheel selection]]
* [[Selection strategy]]
* [[Simulated annealing]]
* [[Stochastic universal sampling]]
* [[Termination condition]]
* [[Tour]]
* [[Tournament selection]]
* [[Truncation selection]]
 
== Helpful resources ==
 
* [https://www.analyticsvidhya.com/blog/2017/07/introduction-to-genetic-algorithm/ This article is a good place to get started]
* a good game: https://andymakes.itch.io/combat-genetics
* a better game to help you understand genetic algorithms: https://david-birge-cotte.itch.io/evolution-sandbox


== Standards ==
== Standards ==

Latest revision as of 09:32, 9 December 2021

HL CONTENT: Modeling & Simulation[1]

The genetic algorithm is a method for solving both constrained and unconstrained optimization problems that is based on natural selection, the process that drives biological evolution. The genetic algorithm repeatedly modifies a population of individual solutions. At each step, the genetic algorithm selects individuals at random from the current population to be parents and uses them to produce the children for the next generation. Over successive generations, the population "evolves" toward an optimal solution. You can apply the genetic algorithm to solve a variety of optimization problems that are not well suited for standard optimization algorithms, including problems in which the objective function is discontinuous, nondifferentiable, stochastic, or highly nonlinear. The genetic algorithm can address problems of mixed integer programming, where some components are restricted to be integer-valued.[2]

Two videos to get you started[edit]


Use of genetic algorithms[edit]

The basic pattern of genetic algorithms[edit]

  1. A random set of solutions would be generated on the sample documents
  2. And tested against the human labelling
  3. Best fit solutions retained
  4. New generation created by mutating/crossing
  5. Algorithm repeated
  6. Until a good fit obtained

Terms associated with genetic algorithms[edit]

Helpful resources[edit]

Standards[edit]

  • Outline the use of genetic algorithms.

References[edit]