Discussion of article "Population optimization algorithms: Monkey algorithm (MA)"
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: Monkey algorithm (MA) has been published:
In this article, I will consider the Monkey Algorithm (MA) optimization algorithm. The ability of these animals to overcome difficult obstacles and get to the most inaccessible tree tops formed the basis of the idea of the MA algorithm.
The area explored by the monkeys is a fitness function landscape, so the highest mountain corresponds to the solution of the problem (we consider the problem of global maximization). From its current position, each of the monkeys moves up until it reaches the top of the mountain. The climb process is designed to gradually improve the value of the target function. Then, the monkey makes a series of local jumps in a random direction in the hope of finding a higher mountain, and the upward movement is repeated. After performing a certain number of climbs and local jumps, the monkey believes that it has sufficiently explored the landscape in the vicinity of its initial position.
In order to explore a new area of the search space, the monkey performs a long global jump. The above steps are repeated a specified number of times in the algorithm parameters. The solution of the problem is declared to be the highest of the vertices found by the given population of monkeys. However, the MA spends significant computational time searching for local optimal solutions in the process of climbing. The global jump process can speed up the rate of convergence of the algorithm. The purpose of this process is to force the monkeys to find new search opportunities so as not to fall into the local search. The algorithm has such advantages as a simple structure, relatively high reliability and a good search for local optimal solutions.
MA is a new type of evolutionary algorithm that can solve many complex optimization problems characterized by non-linearity, non-differentiability and high dimensionality. The difference from other algorithms is that the time spent by the MA is mainly due to the use of the climb process to find local optimal solutions. In the next section, I will describe the main components of the algorithm, the presented solutions, the initialization, the climb, the observation and the jump.
Author: Andrey Dik