NN representation

这一课主要是讲3层神经网络

  

  

下面是常见的 activation 函数.sigmoid, tanh, ReLU, leaky ReLU.

  

Sigmoid 只用在输出0/1 时候的output layer, 其他情况基本不用,因为tanh 总是比sigmoid 好.

两种 ReLU 使用起来总是要比sigmoid 和 tanh 快。ReLU 是最常用的 activation.

  

为什么Activation function 要是non-linear的?因为如下图所示如果activation 是linear的,那么最终output 只是 input 的线性函数.

   

Gradient of activation function

  

  

  

  

Gredient of 2 layer NN.

  

  

Random initialization

  

Coursera, Deep Learning 1, Neural Networks and Deep Learning - week3, Neural Networks Basics的更多相关文章

  1. 【DeepLearning学习笔记】Coursera课程《Neural Networks and Deep Learning》——Week2 Neural Networks Basics课堂笔记

    Coursera课程<Neural Networks and Deep Learning> deeplearning.ai Week2 Neural Networks Basics 2.1 ...

  2. 【DeepLearning学习笔记】Coursera课程《Neural Networks and Deep Learning》——Week1 Introduction to deep learning课堂笔记

    Coursera课程<Neural Networks and Deep Learning> deeplearning.ai Week1 Introduction to deep learn ...

  3. 课程一(Neural Networks and Deep Learning),第四周(Deep Neural Networks) —— 3.Programming Assignments: Deep Neural Network - Application

    Deep Neural Network - Application Congratulations! Welcome to the fourth programming exercise of the ...

  4. 课程一(Neural Networks and Deep Learning),第二周(Basics of Neural Network programming)—— 4、Logistic Regression with a Neural Network mindset

    Logistic Regression with a Neural Network mindset Welcome to the first (required) programming exerci ...

  5. Neural Networks and Deep Learning

    Neural Networks and Deep Learning This is the first course of the deep learning specialization at Co ...

  6. [C3] Andrew Ng - Neural Networks and Deep Learning

    About this Course If you want to break into cutting-edge AI, this course will help you do so. Deep l ...

  7. 《Neural Networks and Deep Learning》课程笔记

    Lesson 1 Neural Network and Deep Learning 这篇文章其实是 Coursera 上吴恩达老师的深度学习专业课程的第一门课程的课程笔记. 参考了其他人的笔记继续归纳 ...

  8. 第四节,Neural Networks and Deep Learning 一书小节(上)

    最近花了半个多月把Mchiael Nielsen所写的Neural Networks and Deep Learning这本书看了一遍,受益匪浅. 该书英文原版地址地址:http://neuralne ...

  9. 课程四(Convolutional Neural Networks),第二 周(Deep convolutional models: case studies) —— 0.Learning Goals

    Learning Goals Understand multiple foundational papers of convolutional neural networks Analyze the ...

  10. 课程一(Neural Networks and Deep Learning),第三周(Shallow neural networks)—— 3.Programming Assignment : Planar data classification with a hidden layer

    Planar data classification with a hidden layer Welcome to the second programming exercise of the dee ...

随机推荐

  1. js jquery radio 选中 选中值

    radio示例: <label><input type="radio" name="type" id="type" val ...

  2. Web Deploy 服务器安装设置与使用

    一.服务器的安装设置 1.在windows server上确保IIS安装了[管理服务]这个功能.方法是在[服务器管理器]=>[管理]=>[添加角色和功能]=>[下一步]=>[基 ...

  3. 怎样删除windows server back 备份副本文件

    我用的服务器是windows server 2012 下面说明 第一步:打开windows powershell 第二步:输入命令   DISKSHADOW 第二步:输入 delete shadows ...

  4. 网络I/O模型总结

    把网络IO模型整理了一下,如下图

  5. 常见的Dos命令

    dir : 列出当前目录下的文件以及文件夹 md : 创建目录 rd : 删除目录    注意:rd不能删除非空的文件夹,而且只能用于删除文件夹. cd : 进入指定目录 cd.. : 退回到上一级目 ...

  6. python自动化开发-[第九天]-异常处理、进程

    今日概要: 1.异常处理使用 2.进程 3.paramiko模块使用 一.异常处理 1.常见的错误异常 #错误异常一 print(a) #NameError #错误异常二 int('sdadsds') ...

  7. Java NIO系列教程(七) selector原理 Epoll版的Selector

    目录: Reactor(反应堆)和Proactor(前摄器) <I/O模型之三:两种高性能 I/O 设计模式 Reactor 和 Proactor> <[转]第8章 前摄器(Proa ...

  8. python 面向对象(六)MRO C3算法 super

    ########################总结################ 面向对象回顾 类:对某一个事物的描述,对一些属性和方法的归类 class 类名: var=123#类变量 def ...

  9. SpringMVC+Shiro不拦截静态资源配置

    最近在弄SpringMVC与Shiro整合,发现如果将DispatcherServlet拦截 *.do这样的URL,就不存在访问不到静态资源的问题.如果DispatcherServlet改为拦截“/” ...

  10. Hadoop记录-变更

    1.安装salt-minion sed -i 's/^#//g' /etc/yum.repos.d/centos7.4.repo sed -i 's/enabled=0/enabled=1/g' /e ...