分类 DL学习笔记 下的文章
[转载]受限玻尔兹曼机及matlab代码

View more

[转载]受限玻尔兹曼机及matlab代码

能量模型的概念从统计力学中得来,它描述着整个系统的某种状态,系统越有序,系统能量波动越小,趋近于平衡状态,系统越无序,能量波动越大。例如:一个孤立的物体,其内部各处的温...
[转载]高斯混合模型GMM

View more

[转载]高斯混合模型GMM

高斯混合模型 本文就高斯混合模型(GMM,Gaussian Mixture Model)参数如何确立这个问题,详细讲解期望最大化(EM,Expectation Maximization)算法的实施过程。 ...
[转载]特征值、特征向量和PCA的概念及理解

View more

[转载]特征值、特征向量和PCA的概念及理解

特征值分解和主成份分析 §协方差矩阵 协方差(Covariance)用于衡量两个变量的总体误差。设两个随机变量X和Y的期望值分别为E(X)和E(Y),则其协方差定义为: ...
[转载]浅谈协方差矩阵的概念及意义

View more

[转载]浅谈协方差矩阵的概念及意义

基于数学基础太差,学习神经网络需要大量的数学知识,而我总是纠结于概率论、矩阵、线性代数等相关知识,只能一点一点积累了。 ...
Coursera: Neural Networks for Machine Learning- Lecture 14

View more

Coursera: Neural Networks for Machine Learning- Lecture 14

DBN(信息量较大)

Coursera: Neural Networks for Machine Learning- Lecture 13

View more

Coursera: Neural Networks for Machine Learning- Lecture 13

Belief Nets & wake-sleep algorithm

Coursera: Neural Networks for Machine Learning- Lecture 12

View more

Coursera: Neural Networks for Machine Learning- Lecture 12

2012_An efficient learning procedure for deep Boltzmann Machine ICML2008_Training restricted Boltzmann machines using approximations to the likelihood gradient(PCD) ...
Coursera: Neural Networks for Machine Learning- Lecture 11

View more

Coursera: Neural Networks for Machine Learning- Lecture 11

Energy-based Models

Coursera: Neural Networks for Machine Learning- Lecture 10

View more

Coursera: Neural Networks for Machine Learning- Lecture 10

Ways to improve generalization

Coursera: Neural Networks for Machine Learning- Lecture 9

View more

Coursera: Neural Networks for Machine Learning- Lecture 9

Ways to improve generalization