Anda di halaman 1dari 2

-

GENETIC ALGORITHM FOR OPTIMIZATION What are genetic algorithms? They have biological origins, and they belong to the larger class of evolutionary algorithms, which generate solutions to optimization problems based on natural evolution. For example, inheritance, mutation, and selection are factors that contribute to the development of these algorithms. Although they are modelled after natural processes, we design our own mutations and selection criteria. Applications: used in several fields such as engineering, economics, chemistry, phylogenetics, and many more Genetic algorithms are a popular strategy to optimize non-linear systems with a large number of variables

Certain definitions are required, so just to go over a few: - parameter: variable - gene: encoded form of parameter. Chromosome: For genetic algorithms, the chromosomes represent a set of genes that code the independent variable. Each chromosome is a solution to the problem we are trying to optimize. These genes when used in general optimization could also be integers, or Boolean, or floating point variables, or a combination of these. Selection: essentially in nature this is determined by how adapted an individual is adapted to their environment. The more adapted, the greater chance it has of surviving and producing offspring with the same genes. We probably know this as survival of the fittest -fitness : is the value we are trying to maximize -recombination: reproduction process, where genes of the parent are used to form a new chromosome. Typical recombination involves 2 parents, but more parents are also possible. Two of the most widely used genetic algorithms: Conventional Crossover and Blending Crossover

Why use genetic algorithms? Genetic algorithms are one of the best ways to solve a problem for which little is known. Theyre relatively general, so they work well in almost any search space. Genetic algorithms have increasingly been applied in the field of economics. They are used to characterize a variety of models, including the cobweb model, game theory, and asset pricing, to name a few.

The most common type of genetic algorithm works in this manner: A population is created randomly with different individuals. These individuals are then evaluated and given a score based on how well they perform on a given task. From this, 2 individuals are selected. The higher their fitness, the greater their chance of being selected. These individuals then reproduce to create one or more offspring, which may be randomly mutated. There are different types of selection. However, one of the more common types is roulette wheel selection. Here, the probably of being selected is directly proportional to an individuals fitness. This process continues until a suitable solution is found, or a certain number of generations has passed depending on the problem. Here is a very simple example to demonstrate the use of genetic algorithms. Problem: Maximize the function f(x) =x2, where 0<x<31 Now obviously, we can tell by simply looking at this that the function will be maximized when x=31. However, this simple example can also be used to illustrate the use of the algorithm. To use genetic algorithms for optimization: 1) 2) 3) 4) 5) choose the parameters we want to optimize determine chromosomal representation of the parameters generate initial population of individuals evaluate the fitness of each individual allow selection rules to select the next population

Applications: economics - game theory game theory is the study of multi-person decision problems. Oligopoly

Anda mungkin juga menyukai