三层前馈人工神经网络研究

Research on Three-layer Feed-forward Artificial Neural Network

人工神经网络的应用越来越广泛,在分析了神经网络的特点和应用趋势的基础上,针对有反馈的神经网络设计,对系统原理进行了简要介绍,利用计算机工具对三层神经网络进行了设计及仿真,仿真结果表明人工神经网络具有非线性自适应的信息处理能力,当隐含层的神经元激活函数采用S形函数,而且隐含层的神经元个数足够多时,通过调整网络的加权系数和阈值,该网络可以逼近任意的非线性函数,但网络结构的确定以及训练样本的取得仍然是建立神经网络的关键。

The artificial neural network is used more and more widely. This paper analyzes the characteristics and application trend of neural network, and introduces briefly the system principle aiming at the neural network with feedback design. The design and simulation of three-layer neural network by using computer tools for three-layer neural network for the design and simulation, and the simulation results show that artificial neural network has nonlinear adaptive information processing capability. When the activation function of hidden layer neurons uses sigmoidal function and the number of hidden layer neurons is enough, the network can approximate any nonlinear function by adjusting the weighted coefficient and threshold of network. And the determination of the network structure and the acquisition of training sample is still the key to estabhsh a neural network.