回归分析

标签：max   ssg   lin   lead   整理   infoq   log   目标   iss   

#学习笔记 待整理
http://itindex.net/detail/50531-%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0

###回归分析能做什么
http://www.advancedtechnic.com/ud/
https://zh.wikipedia.org/wiki/%E8%BF%B4%E6%AD%B8%E5%88%86%E6%9E%90

####线性回归
拟合和预测问题
https://zh.wikipedia.org/wiki/%E7%B7%9A%E6%80%A7%E5%9B%9E%E6%AD%B8
http://blog.csdn.net/xbinworld/article/details/43919445
http://open.163.com/movie/2011/6/A/8/M82IC6GQU_M83JD6PA8.html
http://www.jianshu.com/p/fcd220697182
http://blog.csdn.net/xuelabizp/article/details/50531972
https://www.zhihu.com/question/24900876

http://blog.csdn.net/just_do_it_123/article/details/51056260
http://www.cnblogs.com/Sinte-Beuve/tag/%E6%A2%AF%E5%BA%A6%E4%B8%8B%E9%99%8D/
http://www.solinx.co/archives/758
http://www.solinx.co/archives/721
http://www.solinx.co/archives/800
http://blog.csdn.net/xidianzhimeng/article/details/20847289
https://www.zhihu.com/question/20447622

#### 逻辑回归
实际上是个分类问题
https://chenrudan.github.io/blog/2016/01/09/logisticregression.html#4
http://www.infoq.com/cn/articles/click-through-rate-prediction
http://wenjunoy.com/2016/01/logistic-sigmoid-function.html
http://ziyuanjun.github.io/2016/01/21/Logistic%E5%9B%9E%E5%BD%92%E7%9A%84%E6%A2%AF%E5%BA%A6%E4%B8%8B%E9%99%8D%E6%B3%95%E6%8E%A8%E5%AF%BC/
http://blog.csdn.net/fishmemory/article/details/51598062
http://blog.csdn.net/Fishmemory/article/details/51603836
http://www.cnblogs.com/richqian/p/4511557.html
http://blog.csdn.net/zouxy09/article/details/20319673
http://blog.csdn.net/ligang_csdn/article/details/53838743
http://www.hanlongfei.com/%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0/2015/08/05/mle/
https://tech.meituan.com/intro_to_logistic_regression.html
http://www.cnblogs.com/jerrylead/archive/2011/03/05/1971867.html
https://chenrudan.github.io/blog/2016/01/09/logisticregression.html#4
https://vnicl.net/2017/05/04/Logistic%E5%9B%9E%E5%BD%92/
http://www.cnblogs.com/jerrylead/archive/2011/03/05/1971867.html

#### 逻辑回归和softmax回归的差别
http://blog.csdn.net/xbinworld/article/details/45291009
http://blog.csdn.net/danieljianfeng/article/details/41901063

#### 常用的 损失函数和目标函数 有那些
http://www.cnblogs.com/Belter/p/6653773.html

##### 为什么交叉熵比平均差 代价函数更好
http://blog.csdn.net/u014313009/article/details/51043064
https://www.aiboy.pub/2017/07/16/Cross_Entropy_Cost_Function/
http://blog.csdn.net/haolexiao/article/details/70142571
http://blog.csdn.net/jasonzzj/article/details/52017438
http://blog.csdn.net/behamcheung/article/details/71911133
http://blog.csdn.net/u012162613/article/details/44239919

#### 构造目标函数的方式: 最小二乘法,最大似然法,最大熵模型

#### 求解目标函数的工具: 梯度下降 牛顿法
https://ctmakro.github.io/site/on_learning/gd.html
http://kissg.me/2017/07/23/gradient-descent/
http://blog.csdn.net/lilyth_lilyth/article/details/8973972
https://www.jiqizhixin.com/articles/2016-11-21-4

#### 其他回归方法
http://blog.csdn.net/xbinworld/article/details/44276389

#### 正则化的作用
http://xudongyang.coding.me/regularization-in-deep-learning/
http://hpzhao.com/2017/03/29/%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E4%B8%AD%E7%9A%84%E6%AD%A3%E5%88%99%E5%8C%96/
http://blog.csdn.net/zouxy09/article/details/24971995
http://blog.csdn.net/jinping_shi/article/details/52433975
https://plushunter.github.io/2017/07/22/%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E7%AE%97%E6%B3%95%E7%B3%BB%E5%88%97%EF%BC%8828%EF%BC%89%EF%BC%9AL1%E3%80%81L2%E6%AD%A3%E5%88%99%E5%8C%96/
http://www.jianshu.com/p/a47c46153326
https://vimsky.com/article/969.html
https://liam0205.me/2017/03/30/L1-and-L2-regularizer/
http://t.hengwei.me/post/%E6%B5%85%E8%B0%88l0l1l2%E8%8C%83%E6%95%B0%E5%8F%8A%E5%85%B6%E5%BA%94%E7%94%A8.html

#### 激活函数的作用与分类
http://blog.csdn.net/cyh_24/article/details/50593400
https://www.zhihu.com/question/22334626
http://www.jianshu.com/p/22d9720dbf1a
http://www.cnblogs.com/rgvb178/p/6055213.html
https://www.jiqizhixin.com/articles/2017-10-10-3

#### 最大熵模型 交叉熵
http://blog.csdn.net/v_july_v/article/details/40508465

回归分析

标签：max   ssg   lin   lead   整理   infoq   log   目标   iss   

原文地址：http://www.cnblogs.com/xuxm2007/p/7818052.html