Scholarlink

The simplest and most widely used learning algorithm is the least mean square (LMS) algorithm. This algorithm was used to train ADALINE (adaptive linear neuron). ADALINE was a single trainable neuron. Input training patterns were presented, and the weights (coefficients) were adjusted to deliver neuron outputs that are the best least square fit to the desired outputs of the training patterns. LMS is based on the method of steepest descent. ADALINE is the basic building block of present-day neural networks.

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

Adaline

  • 章节

  • 发表日期：

    2025年

查看全文
  • Bernard Widrow,
  • Edward P. Katz

摘要

The simplest and most widely used learning algorithm is the least mean square (LMS) algorithm. This algorithm was used to train ADALINE (adaptive linear neuron). ADALINE was a single trainable neuron. Input training patterns were presented, and the weights (coefficients) were adjusted to deliver neuron outputs that are the best least square fit to the desired outputs of the training patterns. LMS is based on the method of steepest descent. ADALINE is the basic building block of present-day neural networks.