Roulette wheel selection: Difference between revisions
No edit summary |
No edit summary |
||
| Line 4: | Line 4: | ||
Also known as Fitness Proportionate Selection, Roulette Wheel Selection is a stochastic, or random, selection process to pick out individuals to put into the mating pool, where the likelihood of an individual being selected is proportional to its fitness value in the current generation. Because the sum of all probabilities must be one, all fitness values will be normalised to be between 0 and 1 before selection begins. The following formula calculates the probability an individual is selected, where <math>f</math> denotes fitness, <math>N</math> denotes population, and <math>p</math> denotes probability. | Also known as Fitness Proportionate Selection, Roulette Wheel Selection is a stochastic, or random, selection process to pick out individuals to put into the mating pool, where the likelihood of an individual being selected is proportional to its fitness value in the current generation. Because the sum of all probabilities must be one, all fitness values will be normalised to be between 0 and 1 before selection begins. The following formula calculates the probability an individual is selected, where <math>f</math> denotes fitness, <math>N</math> denotes population, and <math>p</math> denotes probability. | ||
<math> | <math> p_x = \frac{f_x}{\sum_{i=1}^{N} f_i}</math> | ||
Examples: | Examples: | ||
Latest revision as of 20:51, 13 January 2022
Advanced programming[1]
Also known as Fitness Proportionate Selection, Roulette Wheel Selection is a stochastic, or random, selection process to pick out individuals to put into the mating pool, where the likelihood of an individual being selected is proportional to its fitness value in the current generation. Because the sum of all probabilities must be one, all fitness values will be normalised to be between 0 and 1 before selection begins. The following formula calculates the probability an individual is selected, where denotes fitness, denotes population, and denotes probability.
Examples:
- An individual with a fitness value of 10 will be twice as likely to be selected than an individual with a fitness value of 5
- The following image explains it well:
