Artificial intelligence global optimization method based on QUATRE architecture

Abstract

The invention discloses an artificial intelligence global optimization method based on the QUATRE architecture, including: S1, establishing a quasi-affine transformation evolution architecture; S2, setting an evaluation function of the quasi-affine transformation evolution architecture, a global optimal value, an iteration threshold, and calling The maximum number of evaluation functions and the conditions for iteration, and initialize the evolutionary architecture; S3, judge whether the conditions for iteration are met, if yes, continue to perform the following steps, if not, stop; S4, calculate the variation matrix B, and According to the quasi-affine transformation evolution framework, the next-generation particle population XG+1 of the particle population XG is obtained; S5, updating the number of calls to the evaluation function and the next-generation particle population; S6, obtaining the Boolean evolution matrix M and the associated evolution matrix from the evaluation function The optimal value is the current optimal particle of the particle population in G+1 iterations and G=G+1, then return to step S3 for judgment. The invention has few parameters and high efficiency.

Description

technical field [0001] The invention belongs to the field of artificial intelligence, in particular to an artificial intelligence global optimization method based on a QUATRE framework. Background technique [0002] As a branch of artificial intelligence, computational intelligence originated in the middle of the 20th century. Its main idea is to study complex data, observe experimental progress, and solve complex problems that cannot be solved by traditional mathematical formulas and models in reality. . Fuzzy logic, artificial neural network, memetic computing, evolutionary computing, quantum computing, etc. all belong to this field. [0003] Among them, in the field of evolutionary computing, there are mainly the following branches: genetic methods, learning classification systems, evolutionary programming, evolutionary strategies, and swarm intelligence. Among them, various methods in the swarm intelligence branch under the field of evolutionary computing solve complex...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the above technical problems, the purpose of the present invention is to propose a new evolutionary architecture under the branch of swarm intelligence in the field of artificial intelligence, which can not only achieve the purpose of global optimization by realizing the uniform search of the "solution space", thereby eliminating the difference The defects in the evolution method and the particle cluster method can also reduce the implementation complexity of this method in practical applications by setting fewer parameters

Images