Scholarlink

A neural network is based on a simplified model of neuron connected to each other. Each neuron receives inputs and computes an output according to a function defined for each neuron (Fig. 1.1). The function f can be a sigmoid (exponential or tangential), or a Gaussian function, or the sign function. A network of neurons can therefore be represented by the synaptic weights (w) of the different neurons.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Methodology to Build an Artificial Neural Network for Reservoir Engineering Problems

  • 章节

  • 发表日期：

    2025年

查看全文
  • Dominique Guérillot

摘要

A neural network is based on a simplified model of neuron connected to each other. Each neuron receives inputs and computes an output according to a function defined for each neuron (Fig. 1.1). The function f can be a sigmoid (exponential or tangential), or a Gaussian function, or the sign function. A network of neurons can therefore be represented by the synaptic weights (w) of the different neurons.