BP neural network note
来源:互联网 发布:蓝光原盘播放软件tmt 编辑:程序博客网 时间:2024/05/19 05:02
Note系列以学习笔记为主,可能会很零散,主要跟我的需求相关。鉴于我的写作水平,不论中英文,都很差。我就怎么舒服怎么写啦。
Today, I read the paper from Rumelhart et al. and their report. Here I write down some important points and statements. Most of them are words in the paper.
1. They generated a gradient descent method for finding weights in any feedforward network with semilinear units. And they got some further observations:
Any number of weights in the network can be fixed.
Some units recieve two different kinds of error: that from the direct comparison with the target and that passed through the other output units whose activation it affects.
2. I think the most important part is learning procedure, which could evolve a set of weights to produce an arbitrary mapping from input to output. So the essential part isweight.
3. The gradient descent procedure is bound by the problem of local maxima or minima. It's a common problem for any hill climbing procedure.
(From: https://en.wikipedia.org/wiki/Maxima_and_minima#/media/File:Extrema_example_original.svg)
4. A useful activation function. The linear threshold function is discontinuous and hence will not suffice for the generalized delta rule. Thus, we need a continuous, nonlinear activation function. Most are logistic activation function in their experiments. In a practical learning situation in which the desired outputs are binary {0,1}, the system can never actually achieve these values. Therefore, we typically use the values of 0.1 and 0.9 as the targets.
5. The learning rate. The constant of proportionality is the learning rate. The larger this constant, the larger the changes in the weights.
- BP neural network note
- BP Neural Network
- BP Neural Network
- Principle of Bp Neural network
- 【学习笔记3】Neural Network --BP方法
- Threshold function with BP neural network
- MLP && BP && RBF && SVM Neural Network
- Neural Network 反向传播算法(BP Algorithms)推导,无代码
- CalTech machiine learning, video 10 note(Neural Network)
- review CalTech machine learning, video 10 note(Neural Network)
- READING NOTE:LCNN: Lookup-based Convolutional Neural Network
- Paper Note - Learning to Hash with Binary Deep Neural Network
- Neural Network
- Neural Network
- Neural Network
- 深度学习框架Keras学习系列(二):神经网络与BP算法(Neural Network and BP Algorithm)
- Introduction to neural network —— C语言实现BP神经网络识别数字
- Neural Network based on Eorr Back Propagation典型BP网络c++实现
- 【J2EE】——十三个规则
- 【Leetcode】Top K Frequent Elements Python实现
- mysql表数据以本地文件方式导入Hive
- 【Socket编程】篇六之IO多路复用——select、poll、epoll
- sql多表查询之分组查询
- BP neural network note
- java设计模式-适配器模式
- Swift开篇005->闭包
- Shell脚本8种字符串截取方法总结
- CSS选择器
- Asp.Net使用Easy UI DataGrid 实现行过滤[行模糊查询数据]
- eclipse 字体大小 字体颜色
- 前端经验
- Java开发中的23种设计模式详解(转)