Scholarlink

A valuable introduction to neural networks can be found in [1], with numerous variants revisited in the next sections, enabling the construction of nonlinear regressions operating on different typologies of data (lists, images, graphs, time series, ...). Deep-learning is based on the use of neural networks, NN, with several layers (hence the adjective “deep”). Roughly speaking, NNs take some input data and generate an output, after transiting throughout the different neurons of the different layers. To describe NNs in more detail, take a look to the sketch in Fig. 20.1. This simple (the simplest, in fact) artificial neuron receives two input data, \(x_1\) and \(x_2\) , and produces an output y.

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

Neural Network Based Machine Learning Techniques

  • 章节

  • 出版模式：

    开放获取

  • 发表日期：

    2025年

查看全文
  • Francisco Chinesta,
  • Elías Cueto,
  • Victor Champaney,
  • Chady Ghnatios,
  • Amine Ammar,
  • Nicolas Hascoët,
  • David González,
  • Icíar Alfaro,
  • Daniele Di Lorenzo,
  • Angelo Pasquale,
  • Dominique Baillargeat

摘要

A valuable introduction to neural networks can be found in [1], with numerous variants revisited in the next sections, enabling the construction of nonlinear regressions operating on different typologies of data (lists, images, graphs, time series, ...). Deep-learning is based on the use of neural networks, NN, with several layers (hence the adjective “deep”). Roughly speaking, NNs take some input data and generate an output, after transiting throughout the different neurons of the different layers. To describe NNs in more detail, take a look to the sketch in Fig. 20.1. This simple (the simplest, in fact) artificial neuron receives two input data, \(x_1\) and \(x_2\) , and produces an output y.