Network Based on Improved Genetic Algorithms
Download as PDF
DOI: 10.23977/icmmct.2019.62049
Corresponding Author
Qi Anzhi
ABSTRACT
BP Neural Network is a kind of non-linear system which simulates brain information processing algorithm. It has strong distributed information storage, parallel processing and adaptive learning ability. It is a multi-layer feed forward network trained by error back propagation algorithm. Combining it with genetic algorithm, a learning and training approach with good global optimization search and local time-frequency characteristics can be obtained. In this paper, the crossover rate of genetic algorithm is improved, and the efficiency of the algorithm is improved. At the same time, in order to solve the problem that the initial population is far away from the optimal individual, the convergence speed is slow and it is easy to fall into the local minimum. The information gain of the feature is used as the degree of discrimination of each feature to the category. On this basis, it is used to optimize the BP neural network input layer neuron screening method. The experimental results show that the efficiency of the traditional method is significantly improved.
KEYWORDS
Improved genetic algorithm, BP neural network, Neuron