Global optimization, search and machine learning methods based on genetic algorithm

摘要

A global optimization, search and machine learning method based on genetic algorithm is provided. SOLUTION: A global optimization, search and machine learning method based on the genetic algorithm of the present invention comprises the step 1 based on the objective function f (x) of the structure of the problem substance and the problem substance as a chromosome of the genetic algorithm. Based on Step 2 which is edited and the operation parameters are automatically calculated or input and initialized, and the evaluation for the objective function f (x) Step 3 in which optimization is repeatedly performed on the current generation (k generation) group Gk = {P1k, P2k,..., PSk} by the operator, and a step 4 in which the optimal solution of the problem object is output And. The present invention is a natural method of "heriting of acquired traits" and "Use-and-Disuse Operator" of Lamarck's theory of evolution of natural methods of modern "epigenetics" and of "suitable person survival" of Darwinian theory of evolution. Combine within the range and simplify the genetic algorithm structure. Overcome many technical flaws present in conventional algorithms, increase their efficiency, improve global optimization and late evolution sustainability, and more global optimization, search and machine learning problems A more favorable solution is obtained. [Selected figure] Figure 1