An Evolutionary Algorithm Based Adaptive Learning Neural Network Realization Method

Abstract

The invention belongs to the technical field of neural network computing, and is an evolutionary algorithm-based self-adaptive learning neural network implementation method, using one or several known neural networks as the initial parent of the evolutionary algorithm, and integrating the evolutionary algorithm as the first generation of the evolutionary algorithm. Describe the characteristics of each neural network of the initial parent generation, so as to obtain the optimal output value. The present invention performs binary coding on the circuit realized by the neural network, and uses the result obtained by the coding as an individual chromosome, and each chromosome constitutes the biological body. The original population, that is, the initial parent generation, the present invention breaks through the previous situation of only using the evolutionary algorithm to optimize the neural network weights, and realizes the simultaneous optimization of the neural network organization form, the connection weights between the networks, and the network calculation method by using the evolutionary algorithm. Carry out optimization, enhance the degree of freedom of the network, and expand the scope of optimization; initially obtain a relatively simple network, and increase the complexity of the network through algorithms in the acquired learning.

Description

technical field [0001] The invention belongs to the technical field of neural network calculations, and relates to the realization of various learnable neural network structures and the optimization of evolutionary algorithms in network calculations, in particular to a method for realizing self-adaptive learning neural networks based on evolutionary algorithms. Background technique [0002] The current research on neural networks can be roughly divided into three mainstream structural models: Perceptron neural network, Back-propagate neural network and Deep Learning neural network. Each network has its own characteristics. In terms of overall performance, the deep learning network performs better than the other two in image recognition. [0003] The perceptual neural network only has a three-layer structure. The first layer usually abstracts the characteristics of the target artificially. The second layer is a computing network. The weight of the features abstracted by the f...

AI Technical Summary

Problems solved by technology

However, in terms of the implementation structure, the complexity of the deep learning network is relatively high, and the recognition accuracy has a great relationship with the number of training

Images