Fleet 3.11.0 released with software inventory

Fleet 3.11.0 is now available and more powerful, with improved performance, support for software inventory, and enhancements to running common queries. Fleet is utilized by organizations with up to…

Smartphone

独家优惠奖金 100% 高达 1 BTC + 180 免费旋转




Local Versus Global Reinsertion

Genetic Algorithms in Elixir — by Sean Moriarity (68 / 101)

👈 Growing and Shrinking Populations | TOC | What You Learned 👉

One concept you may come across when working with genetic algorithms is the concept of local populations. Local populations are populations that consist of neighborhoods of chromosomes. A neighborhood is a collection of adjacent chromosomes in a population. Neighborhoods are of different types with various structures. The following image illustrates a basic linear neighborhood:

images/ReplacingAndTransitioning/LinearNeighborhood.png

Local populations constrain interactions between chromosomes. A chromosome can only interact with chromosomes inside its neighborhood. Additionally, in some local populations, chromosomes can migrate between neighborhoods. Some genetic algorithms employ special techniques like assigning neighborhood leaders.

Genetic algorithms that employ local populations are often called multi-population genetic algorithms. Strategies for implementing multi-population genetic algorithms fall out of the scope of this book; however, it’s useful to understand what they are and how they differ from the genetic algorithms you’ve implemented.

Because chromosomes are constrained to only interact with other chromosomes in their neighborhood, multi-population genetic algorithms require special selection and reinsertion strategies for ensuring neighborhoods maintain their local integrity. This is where the notion of global versus local reinsertion comes into play. Local reinsertion operates on neighborhoods; global reinsertion operates on entire populations.

Multi-population genetic algorithms are often used to help parallelize a genetic algorithm because evolution can take place independently in each neighborhood. Multi-population genetic algorithms are also used to simulate competition over resources in different environments.

👈 Growing and Shrinking Populations | TOC | What You Learned 👉

Add a comment

Related posts:

Stay present with your ideas!

This is my first article in medium! I have been waiting to perfect the goals and essence and the direction I want to go with my writing for so long that I got tired of my own perfectionism. The main…

The Writer Who Criticized My Writing Openly

Work on your writing. Your words and sentences are awkward. This sentence doesn’t make sense. I would write this sentence this way. I heard these words fro