Discussing the article: "Population optimization algorithms: Stochastic Diffusion Search (SDS)"
So far, I've had little understanding of the implementation. So my question is.
If only half of the input parameters should be optimised in FF, is it possible to ask to optimise ostensibly all parameters anyway, but only the "unnecessary" half should have an unchangeable range?
So far, I've had little understanding of the implementation. So here's a question.
If only half of the input parameters should be optimised in FF, is it possible to ask to optimise ostensibly all parameters anyway, but only the "unnecessary" half should have a fixed range?
Yes, of course, set the range of the corresponding parameters MIN=MAX (the step will not play a role, you can set any), if I understood the question correctly. In this case these parameters will be output with the same value.
In general, the algorithm does not cause calculation of FF, so it can only optionally display the parameters needed by the user.
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
Check out the new article: Population optimization algorithms: Stochastic Diffusion Search (SDS).
The article discusses Stochastic Diffusion Search (SDS), which is a very powerful and efficient optimization algorithm based on the principles of random walk. The algorithm allows finding optimal solutions in complex multidimensional spaces, while featuring a high speed of convergence and the ability to avoid local extrema.
Interesting facts:
1. Stochastic Diffusion Search (SDS) was the first Swarm Intelligence metaheuristic, which belongs to the family of swarm intelligence and natural search and optimization algorithms. Other examples of such algorithms are ant colony optimization, particle swarm optimization and genetic algorithms.
2. Unlike ant colony optimization based on stigmergy communication, SDS uses direct communication between agents, similar to the tandem calling mechanism used by Leptothorax acervorum ants.
The SDS algorithm is based on low-cost partial evaluation of a hypothesis (a candidate solution to a search problem) by agents. The agents then exchange information about the hypotheses through direct personal communication. Through the diffusion mechanism, high-quality solutions can be identified from clusters of agents with the same hypothesis.
Gold mining game
A group of friends, consisting of experienced miners, learn about the possibility of mining gold in the hills of a mountain range. However, they have no information about where exactly the richest place is located. On their maps, the mountain range is divided into several individual hills, each containing a set of strata that require mining. The probability of discovering gold over time is proportional to its wealth.
To maximize their collective wealth, miners should identify the hill with the richest seams of gold so that the maximum number of miners can mine there. However, this information is not available in advance. To solve this problem, miners decide to use simple stochastic diffusion search.
The mining process begins with each miner being randomly assigned a hill to mine (custom hill hypothesis). Every day, each miner randomly selects a seam on their hill to mine.
Author: Andrey Dik