Evolutional Design and Training Algorithm for Feedforward Neural Networks
Evolutional Design and Training Algorithm for Feedforward Neural Networks
复制标题
前馈神经网络的进化设计和训练算法
DOI：
--
复制
发表时间：
1999
影响因子：
0.7
通讯作者：
M. Nakajima
中科院分区：
文献类型：
--
作者：
Haruhisa Takahashi;M. Nakajima
在使用神经网络的模式识别中，研究人员或用户很难为特定任务设计最佳的神经网络架构。任何类型的神经网络架构都有可能获得一定程度的识别率。然而，从识别率和训练有效性上分析，很难获得针对特定任务的最佳神经网络架构。本文提出了一种训练和设计前馈神经网络的进化方法。在所提出的方法中，神经网络被定义为一个个体和其架构与一个物种相同的神经网络。这些网络通过归一化 M.S.E.（均方误差）进行评估，它呈现了网络训练模式的性能。然后，它们的架构根据此处提出的演化规则进行演化。神经网络的架构，换句话说，物种，是通过与个体标准相比的另一种标准测量来评估的。该标准评估物种中最优秀的个体以及物种的进化速度。根据标准，物种的种群规模会增加或减少。进化规则产生了与高级物种略有不同的神经网络架构。因此，所提出的方法可以生成各种神经网络架构。介绍了执行简单 3 × 3 和 4 × 4 像素的神经网络的设计和训练，其中包括垂直、水平和斜线分类以及手写片假名识别。还讨论了所提出方法的效率。关键词: 设计神经网络, 训练神经网络, 前馈神经网络, 手写字符识别, 进化计算
In pattern recognition using neural networks, it is very difficult for researchers or users to design optimal neural network architecture for a specific task. It is possible for any kinds of neural network architectures to obtain a certain measure of recognition ratio. It is, however, difficult to get an optimal neural network architecture for a specific task analytically in the recognition ratio and effectiveness of training. In this paper, an evolutional method of training and designing feedforward neural networks is proposed. In the proposed method, a neural network is defined as one individual and neural networks whose architectures are same as one species. These networks are evaluated by normalized M.S.E.(Mean Square Error) which presents a performance of a network for training patterns. Then, their architectures evolve according to an evolution rule proposed here. Architectures of neural networks, in other words, species, are evaluated by another measurement of criteria compared with the criteria of individuals. The criteria assess the most superior individual in the species and the speed of evolution of the species. The species are increased or decreased in population size according to the criteria. The evolution rule generates a little bit different architectures of neural network from superior species. The proposed method, therefore, can generate variety of architectures of neural networks. The designing and training neural networks which performs simple 3 × 3 and 4 × 4 pixels which include vertical, horizontal and oblique lines classifications and Handwritten KATAKANA recognitions are presented. The efficiency of proposed method is also discussed. key words: designing neural networks, training neural networks, feedforward neural network, handwritten character recognition, evolutional computation