Discussion of article "Population optimization algorithms: Saplings Sowing and Growing up (SSG)"
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
New article Population optimization algorithms: Saplings Sowing and Growing up (SSG) has been published:
Saplings Sowing and Growing up (SSG) algorithm is inspired by one of the most resilient organisms on the planet demonstrating outstanding capability for survival in a wide variety of conditions.
The algorithm is one of the few that does not have a clear description by the authors (only general provisions and ideas are provided). The algorithm operators presented by the authors are also not ready-made instructions for the algorithmic implementation of the program. There are no clear instructions about child and parent trees and their interaction. There are no requirements for the order, in which operators are executed, and any user can change their order to obtain a better seedling.
In a broad sense, SSG is not an optimization algorithm, it is a general set of rules that is designed to complement other algorithms to improve the quality of optimization. In other words, SSG is an add-on for any evolutionary population algorithms, so I have room for imagination and the opportunity to experiment with a specific implementation of the optimization algorithm. I applied some of my own thoughts and experience while writing previous algorithms and used them to work with SSG. The results of the experiments are presented for the reader's judgment below.
To start understanding the algorithm, we need to think of a tree as an optimization agent. A tree is a solution to an optimization problem, where each branch is an optimized parameter of the problem. An abstract and artistic illustration of child and parent trees is provided in Figure 1. The tree trunk is a set of parameters to be optimized. Each branch is a separate optimized parameter, where the length of the branch is limited by the allowable range of values of the corresponding parameter. The direction of the branches does not matter and is only shown in the figure to highlight their difference.
Author: Andrey Dik