houzhicheng.com
机器学习之线性回归
http://houzhicheng.com/ml/2013/04/28/linear-regression-of-machine-regression.html
Apr 28, 2013. 其中 $ theta$ 称为参数或权重, $x$ 为输入或特征,$n$ 为特征个数 不包括$x 0$. 线性回归的优化目标是使cost function最小,其中cost function 用 $J( theta)$ 表示, 为. 这个函数叫做普通最小二乘函数 ordinary least squares。 其中$ theta$开始与初始猜测值,并且更新同时作用于所有参数$j= 0, n$。 这就是所谓的最小二乘法 Least Mean Squares, LMS ,也叫Widrow-Hoff法。 当训练样本多于一个时,由此引出两种方法可以处理多个样本的样本集 批梯度下降法(batch gradient descent)和随机梯度下降法或增量梯度下降法(stochastic gradient descent or incremental gradient descent). Repeat until convergence {. For i=1 to m, {. 其中$X$ 为$m-by-n 1$的矩阵,$(x {(i)}) T$ 为第$i$个训练样本:.
houzhicheng.com
机器学习之logistic回归
http://houzhicheng.com/blog/ml/2014/01/11/machine-learning-logistic_reg.html
Jan 11, 2014. 其中 $ theta$ 称为参数或权重, $x$ 为输入或特征,$n$ 为特征个数 不包括$x 0$. 对于该假设模型的解释为 对于输入$x$,$h(x)$ 为在$ theta$条件下,分类输出为1的概率,数学表示如下: $ h(x) = P(y=1. X; theta) $. Logistic回归的优化目标是使cost function最小,其中cost function 用 $J( theta)$ 表示, 为. 对比发现这和线性回归的形式一模一样,但其实这并不是相同的算法,因为不同于线性回归,logistic 回归的$h(x)$是 $ theta T x$ 的非线性函数。 Logistic 回归的假设模型在$ theta Tx$后加入了一层sigmoid函数,此时$ theta Tx$不作为直接输出而是作为决策边界,对于线性的$ theta Tx$决策边界为一条直线,直线两侧分为两类,对于非线性的$ theta Tx$,决策边界则为非线性的。 Stanford machine learning cs229. Tech blog, interest in big data.
xuyichao.cn
Resources - Yichao Xu
http://www.xuyichao.cn/resources.html
Zouxy09: a list of code and papers for CV, ML. Wu Huaiyu@CAS: all kinds of resources. Computer Vision Industry: a list by David Lowe. Advances in Computer Vision @ MIT. Computer Vision @ U Washington. Machine Learning @ Stanford. Image Understanding II @ UBC. Computational Photography [MIT Raskar]. Foundations of Variational Image Analysis @ Universität Heidelberg. Middlebury: A Database and Evaluation Methodology for Optical Flow. SimpleFlow: A Non-iterative, Sublinear Optical Flow Algorithm.
czinamerica.wordpress.com
The second first quarter | Culture Shock
https://czinamerica.wordpress.com/2014/10/18/the-second-first-quarter
Chuan-Zheng discovers America. (Standard opinion disclaimers apply.). The second first quarter. One year ago today. Drowning in a sea of coursework and ruthlessly dragged along by the high-speed train of graduate school, I wrote, if I may quote myself, that “there must be at least some element of masochism in anyone who voluntarily stays around here.” That suspicion turned out to be correct, perhaps even a mild understatement. The introductory electrical engineering class I’m helping teach this qua...
fdatamining.blogspot.com
F# and Data Mining: October 2014
http://fdatamining.blogspot.com/2014_10_01_archive.html
F# and Data Mining. Saturday, October 4, 2014. A list of references in machine learning and programming languages. Minka, A comparison of numerical optimizers for logistic regression. 160; Estimating a Gamma distribution. 160; Beyond Newton’s method. Steyvers, Multidimensional scaling. Von Luxburg, A tutorial of spectral clustering. Das et. al, Google news personalization: scalable online collaborative filtering. Hofmann, Probabilistic latent semantic analysis. Wickham, Tidy data. 160;...
fdatamining.blogspot.com
F# and Data Mining: A list of references in machine learning and programming languages
http://fdatamining.blogspot.com/2014/10/a-list-of-references-in-machine.html
F# and Data Mining. Saturday, October 4, 2014. A list of references in machine learning and programming languages. Minka, A comparison of numerical optimizers for logistic regression. 160; Estimating a Gamma distribution. 160; Beyond Newton’s method. Steyvers, Multidimensional scaling. Von Luxburg, A tutorial of spectral clustering. Das et. al, Google news personalization: scalable online collaborative filtering. Hofmann, Probabilistic latent semantic analysis. Wickham, Tidy data. 160;...
cspoetry.com
[Stanford]机器学习(CS229) - CS·Poetry
http://cspoetry.com/stanford机器学习cs229.html
课程主页: cs229.stanford.edu. 课程材料下载地址 含讲义,作业题,课程笔记 cs229.stanford.edu/materials.html. 课程视频 网易公开课 : 斯坦福大学公开课 机器学习课程. Hinton的 这个coursera上貌似也有 : Introduction to Neural Networks and Machine Learning. CMU的 注意无中文 : Introduction to Machine Learning. USC的Machine Learning: An introductory course on machine learning. CMU的统计机器学习 此为上面那个的进阶课程 : Statistical Machine Learning. Comments are off this post. Graduate in ML and CG.