Artificial Intelligence - Genetic Algorithms (GA)

In the past few days of reading (artificial intelligence related), I have encountered many strange "concepts". For example, the following genetic algorithm

Genetic Algorithm

Genetic algorithm is a search algorithm used to solve optimization in computational mathematics, and it is a kind of evolutionary algorithm. Evolutionary algorithms were originally developed from some phenomena in evolutionary biology, including heredity, mutation, natural selection, and hybridization . Genetic algorithms are usually implemented as a computer simulation. For an optimization problem, a population of abstract representations (called chromosomes) of a certain number of candidate solutions (called individuals) evolve towards better solutions .

Genetic Algorithm (GA) was created by John Holland of the University of Michigan in 1975, based on Darwin's theory of evolution

Irwin's theory of evolution : Darwin believed that there is a struggle for survival among organisms, the fittest survive, and the unfit are eliminated, which is natural selection. It is through heredity, variation and natural selection that organisms evolve and develop from low to high, from simple to complex, and from less to more .

Furthermore, Darwin believed that "mutation" is random and that "evolution" and progress are different. Accidental variation in no particular direction is a mechanistic one .
Therefore, it is determined that " excellent individual = excellent answer ", and let the computer use the method of evolution to find the best answer , which is the genetic algorithm.

How to use genetic algorithm

Genetic Algorithms are good at finding or solving the best one from the "innumerable answers" .

The process is as follows:

Genetic Algorithm Applications

Since the overall search strategy and optimization search method of the genetic algorithm do not depend on gradient information or other auxiliary knowledge, but only need the objective function and the corresponding fitness function that affect the search direction, the genetic algorithm provides a solution to complex systems. A general framework for a problem, it does not depend on the specific domain of the problem, it is very robust (robust and robust) to the kind of problem, so it is widely used in many sciences.

Genetic algorithms are widely used in games, stock trading, flight route optimization, and aircraft wing size lights.

insufficiency

1. Coding irregularities and coding inaccuracies.
2. A single genetic algorithm code cannot comprehensively express the constraints of the optimization problem. One way to account for constraints is to use a threshold for infeasible solutions, so that the computation time must increase.
3. Genetic algorithms are generally less efficient than other traditional optimization methods.
4. Genetic algorithms are prone to premature convergence.
5. Genetic algorithm has no effective quantitative analysis method for the accuracy, feasibility and computational complexity of the algorithm.

References

1. "Comic Artificial Intelligence"
2. Baidu Encyclopedia - Genetic Algorithm