Genetic Algorithms

Sablier
217.2K views

Open Source Your Knowledge, Become a Contributor

Technology knowledge has to be shared and made accessible for free. Join the movement.

Create Content

Algorithm

Algorithm overview

  1. Create the base population We create a random initial population. Each individual is is defined by its genetic material. We create a new individual with the previously declared create_chromosome(size) function.
  2. Evaluation Each individual is scored on its fitting to the problem. This is done in the beginning of the selection.
  3. Selection Each individual has a chance to be retained proportional to the way it fits the problem. We only keep the selected individuals returned by the selection(population) function.
  4. Crossover / reproduction Random couples are formed in the selected population. Each couple produces a new individual. The number of individuals in the population can either be constant or vary over time.

On each reproduction :

  • Crossover The genetic material of a child is a combination of the parents' (generally 50% of each parent's genetic material). Once the parents have been chosen the ``crossover(parent1, parent2)` function allows the creation of the child.
  • Mutation Probability : from 0.1% to 1% Each child have a chance to have a randomly modified gene thanks to the mutation(chromosome) function.

Finally, the is_answer(chromosome) function checks if the individual is solution to the problem (100% score). If there is no solution, we go to the next generation (phase 2).

Recap

The goal of this exercise is to find the secret sentence using a genetic algorithm and the tools we created earlier.

Genetic algorithm
Open Source Your Knowledge: become a Contributor and help others learn. Create New Content