机器学习（二）学习规划及资料汇总

骑猪看日落 2022-08-02 11:42 229阅读 0赞

\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*

原文链接：http://ml.memect.com/article/machine-learning-guide.html

**基本概念**

机器学习是近20多年兴起的一门多领域交叉学科，涉及**概率论**、**统计学**、**逼近论**、**凸分析**、**算法复杂度理论**等多门学科。机器学习理论主要是设计和分析一些让计算机可以自动“学习”的算法。机器学习算法是一类从数据中自动分析获得规律，并利用规律对未知数据进行预测的算法。因为学习算法中涉及了大量的统计学理论，机器学习与统计推断学联系尤为密切，也被称为统计学习理论。算法设计方面，机器学习理论关注可以实现的，行之有效的学习算法。

下面从微观到宏观试着梳理一下机器学习的范畴：一个具体的算法，领域进一步细分，实战应用场景，与其他领域的关系。

![20150718091935940][]

图1: 机器学习的例子：NLTK监督学习的工作流程图 (source: [http://www.nltk.org/book/ch06.html][http_www.nltk.org_book_ch06.html])

![20150718092108064][]

图2: 机器学习概要图 by Yaser Abu-Mostafa (Caltech) (source: [http://work.caltech.edu/library/181.html][http_work.caltech.edu_library_181.html])

![20150718092055268][]

图3: 机器学习实战：在python scikit learn 中选择机器学习算法 by Nishant Chandra

(source: [http://n-chandra.blogspot.com/2013/01/picking-machine-learning-algorithm.html][http_n-chandra.blogspot.com_2013_01_picking-machine-learning-algorithm.html])

![20150718092156775][]

图4: 机器学习和其他学科的关系： 数据科学的地铁图 by Swami Chandrasekaran (source: [http://nirvacana.com/thoughts/becoming-a-data-scientist/][http_nirvacana.com_thoughts_becoming-a-data-scientist])

**入门攻略**

按照原文的介绍，主要分为三大步骤：1.起步感悟，2.实战笔记，3.行家导读

*  1. [机器学习入门者学习指南 @果壳网][Link 1] (2013) 作者 [白马][Link 2] -- \[起步体悟\] 研究生型入门者的亲身经历。

<table style="font-family:微软雅黑"> 
 <tbody> 
  <tr> 
   <td> 
    <div style="font-family:微软雅黑; font-size:14px; line-height:21px; widows:auto"> 
     <span style="font-family:Arial,Helvetica,sans-serif; font-size:13.3333339691162px">这是一个研究生从刚开始学习机器学习时的起步感悟，有关机器学习的详细介绍，可以参看维基百科</span> 
     <a href="https://en.wikipedia.org/wiki/Machine_learning" style="font-family:'Linux Libertine',Georgia,Times,serif; font-size:12px" rel="nofollow">Machine learning</a> 
     <span style="font-family:'Linux Libertine',Georgia,Times,serif; font-size:12px">，</span> 
     <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; line-height:1.5; background-color:inherit">下面是白马给出的学习方法和步骤，非常感谢白马的指导:</span> 
    </div> 
    <div style="font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> 
     <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">1. 在Coursera上学习Andrew Ng的《机器学习》课程，完成所有的作业，最好能全部拿满分。白马的评论是：<span style="background-color:inherit">老师讲的是深入浅出，不用太担心数学方面的东西。而且作业也非常适合入门者，都是设计好的程序框架，有作业指南，根据作业指南填写该完成的部分就行。这门课上完了，你基本上可以开始简单的应用各种机器学习技术了。</span></span> 
    </div> 
    <div style="font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> 
     <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">2. 独立完成一个小项目，最好再找一本稍微具体一点的关于机器学习的书籍，开始深入研究。可以按照白马的建议找一个开源的项目做做，白马看的书是李航的《统计学习方法》，并且这本书是从自然语言处理的角度来写的，此条的最后白马意见是2和3同时进行比较好，这需要充足的精力和大量的时间。</span> 
    </div> 
    <div style="font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> 
     <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">3. 修炼内功。包括线性代数，矩阵论，概率统计和优化算法等方面，白马原作贴图如下，以备份。并且是机器学习教父级别的人物Jordan推荐的书籍，都是内功秘密，需要认真研读。</span> 
    </div> 
    <div style="font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> 
     <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit"><em><strong>优化</strong></em></span> 
    </div> 
    <div style="font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> 
     <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit"><em>Numerical Optimization(Jorge Nocedal)&nbsp;</em></span> 
    </div> 
    <div style="font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> 
     <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit"><em>&nbsp; &nbsp; &nbsp; &nbsp; csdn下载地址：<u>http://download.csdn.net/detail/xybingo/4120417</u>，没有积分，如果谁有完本，麻烦评论以下，谢谢。</em></span> 
    </div> 
    <div style="font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> 
     <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit"><em>Convex Optimization(boyd)</em></span> 
    </div> 
    <div style="font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> 
     <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit"><em>&nbsp; &nbsp; &nbsp; &nbsp; <span style="font-size:13.3333339691162px; line-height:20px">备份下载地址：<u>http://pan.baidu.com/s/14nFOy</u><span style="font-size:13.3333339691162px; line-height:20px">，</span></span>斯坦福大学原书网站：<u>http://web.stanford.edu/~boyd/cvxbook/</u></em></span> 
    </div> 
    <div style="font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> 
     <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit"><em>Convex Analysis(Rockafellar)</em></span> 
    </div> 
    <div style="font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> 
     <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit"><em>&nbsp; &nbsp; &nbsp; &nbsp; 备份下载地址：<u>http://pan.baidu.com/s/1jGCeLYQ</u></em></span> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>Convex Analysis and Nonlinear Optimizaiton Theory and Examples(Borwein)</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>&nbsp; &nbsp; &nbsp; &nbsp; 没找到下载地址</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em><strong>线性代数，矩阵论，概率统计</strong></em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>Matrix Computation (golub)</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>&nbsp; &nbsp; &nbsp; &nbsp; 备份下载地址：</em> 
     <u><em>http://pan.baidu.com/s/1c0wIarQ</em></u> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>Numerical Linear Algebra(Allaire)</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <span style="font-size:13.3333339691162px; line-height:20px"><em>&nbsp; &nbsp; &nbsp; &nbsp; 没<span style="font-size:13.3333339691162px; line-height:20px">找到下载地址</span>，有普林斯顿大学的学习网站：<u>http://introcs.cs.princeton.edu/java/95linear/</u></em></span> 
     <br> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>Method of Multivariate Analysis(Rencher)</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>&nbsp; &nbsp; &nbsp; &nbsp; 没<span style="font-size:13.3333339691162px; line-height:20px">找到下载地址</span>。</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>Applied Multivariate Statistical Analysis(Hardle)</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em><span style="font-size:13.3333339691162px; line-height:20px">&nbsp; &nbsp; &nbsp; &nbsp; 备份下载地址：<u>http://pan.baidu.com/s/1mgpGdW8</u></span></em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>Matrix Analysis and applied linear algebra, matrix analysis(Rajendra)</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em><span style="font-size:13.3333339691162px; line-height:20px">&nbsp; &nbsp; &nbsp; &nbsp; 没有下载地址，官方网站：<u>http://www.matrixanalysis.com/</u></span></em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>Bayesian theory(Bernardo &amp; Smith)</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em><span style="font-size:13.3333339691162px; line-height:20px">&nbsp; &nbsp; &nbsp; &nbsp;&nbsp;<span style="font-size:13.3333339691162px; line-height:20px"><span style="font-size:13.3333339691162px">没有下载地址</span></span></span></em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>Probalitity and Computing: Randomized Algorithms and Probabilistic Analysis(Mitzenmacher)</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em><span style="font-size:13.3333339691162px; line-height:20px">&nbsp; &nbsp; &nbsp; &nbsp; 备份下载地址：<u>http://pan.baidu.com/s/1nt25s29</u></span></em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>Matrix Differential Calculus With &nbsp;Applications in Statistic and Econometrics(magnus)</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em><span style="font-size:13.3333339691162px; line-height:20px">&nbsp; &nbsp; &nbsp; &nbsp; 没有下载地址</span></em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>Spectral Graph Theory(Chung)</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em><span style="font-size:13.3333339691162px; line-height:20px">&nbsp; &nbsp; &nbsp; &nbsp; 网站链接：<u>http://www.math.ucsd.edu/~fan/research/revised.html</u></span></em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>All of Statistic A Concise Course in Statistical Inference(Wasserman)</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em><span style="font-size:13.3333339691162px; line-height:20px">&nbsp; &nbsp; &nbsp; &nbsp; 网站链接：<u>http://www.stat.cmu.edu/~larry/all-of-statistics/</u></span></em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>Porbability Theory The Logic of Science (jaynes)</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em><span style="font-size:13.3333339691162px; line-height:20px">&nbsp; &nbsp; &nbsp; &nbsp; CSDN英文原版下载地址：<u>http://download.csdn.net/detail/harryxusu/1925470#comment</u>，没有积分这里无法共享，但是可以在线看。</span></em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>数理统计学简史（陈希孺）</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em><span style="font-size:13.3333339691162px; line-height:20px">&nbsp; &nbsp; &nbsp; &nbsp; 备份下载地址：<u>http://pan.baidu.com/s/1i3q2kyp</u></span><br> </em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em><strong>信息论</strong></em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>elements of information theory(cover)</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em><span style="font-size:13.3333339691162px; line-height:20px">&nbsp; &nbsp; &nbsp; &nbsp; 备份下载地址：<u>http://pan.baidu.com/s/1eQ2d8fo</u></span><br> </em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>Information Theory, Inference, and Learning Algorithms(Mackay)</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em><span style="font-size:13.3333339691162px; line-height:20px">&nbsp; &nbsp; &nbsp; &nbsp; 备份下载地址：<u>http://pan.baidu.com/s/1i3EOKIT</u>。</span><br> </em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em><strong>实分析+测度论</strong></em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em>测度论讲义+习题答案（严加安）</em> 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <em><span style="font-size:13.3333339691162px; line-height:20px">&nbsp; &nbsp; &nbsp; &nbsp; 备份下载地址：<u>http://pan.baidu.com/s/1eQ6Nw7C</u>，习题答案：<u>http://pan.baidu.com/s/1c04j7qC</u>。</span><br> </em> 
    </div> 
    <div style="widows:auto; line-height:20px">
      Real Analysis and Probability (R. M. Dudley) 
    </div> 
    <div style="widows:auto; line-height:20px"> 
     <span style="font-size:13.3333339691162px; line-height:20px">&nbsp; &nbsp; &nbsp; &nbsp; 备份下载地址：<u>http://pan.baidu.com/s/1bnnW7Zx</u>。</span> 
     <br> 
    </div> 
    <div style="font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> 
     <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">4. 放弃你的种子和妹子吧。。。</span> 
    </div> </td> 
  </tr> 
 </tbody> 
</table>

*  [有没有做机器学习的哥们？能否介绍一下是如何起步的 @ourcoders][_ourcoders] -- \[起步体悟\] 研究生型入门者的亲身经历，尤其要看[reyoung][]的建议，以下是reyoung的建议：

<table style="font-family:微软雅黑"> 
 <tbody> 
  <tr> 
   <td> <p style="margin-top:5px; margin-bottom:5px; font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">不要看网易的andrew ng公开课了。时间太久远，很多东西都不一样了(这个跟写程序也是一样的，太落后的东西没啥用，除非是经典问题)。看视频的话，看andrew ng在coursera的machine learning视频，14年machine learning的刚要开课，推荐报名，跟着上课，做课后习题和期末考试。(因为只看不干，啥都学不会)。。</span></p> <p style="margin-top:5px; margin-bottom:5px; font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">另外，重点是你要学ML干什么？可选项有如下这些。给出一个选项，可以更针对性的说明应该学什么:</span></p> 
    <div style="font-family:微软雅黑; font-size:14px; line-height:21px; widows:auto"> 
     <div style="background-color:inherit; line-height:20px"> 
      <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">&nbsp;&nbsp;&nbsp;&nbsp;-&nbsp;最快最方便的使用一个通用方法做一个分类器或者回归器(例如验证码识别,etc)</span> 
     </div> 
     <div style="background-color:inherit; line-height:20px"> 
      <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">&nbsp;&nbsp;&nbsp;&nbsp;-&nbsp;从理论上想改进某一种机器学习的学习算法或数据结构(模型)</span> 
     </div> 
     <div style="background-color:inherit; line-height:20px"> 
      <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">&nbsp;&nbsp;&nbsp;&nbsp;-&nbsp;更好的了解各种机器学习算法的特点，应对不同的问题，选择不同的方法。</span> 
     </div> 
     <div style="background-color:inherit; line-height:20px"> 
      <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">&nbsp;&nbsp;&nbsp;&nbsp;-&nbsp;利用现在的硬件产品(显卡，集群)，更好的实现一套机器学习算法。</span> 
     </div> 
    </div> <p style="margin-top:5px; margin-bottom:5px; font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">基本上上面这些都是学Machine learning的一些点，每个方向要学的东西都不一样(虽然基础还是基础)，甚至很多方向不需要学，只需要用一些现成的东西就好。用现成的东西，好处就是速度快，成本小；坏处就是不明白细节。</span></p> 
    <hr style="font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> <p style="margin-top:5px; margin-bottom:5px; font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">只要学习机器学习，一定会看的书籍我推荐一下:</span></p> 
    <div style="font-family:微软雅黑; font-size:14px; line-height:21px; widows:auto"> 
     <div style="background-color:inherit; line-height:20px"> 
      <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">&nbsp;&nbsp;&nbsp;&nbsp;-&nbsp;Mitchell 的《机器学习》。Mitchell是机器学习的鼻祖，第一个提出机器学习概念的人。这本书很薄，很简单。内容很陈旧，但是都是机器学习的经典问题。而且，这本书概念清晰正确(很可贵啊，又简单又正确的书，说明作者功力很强)。</span> 
     </div> 
     <div style="background-color:inherit; line-height:20px"> 
      <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">&nbsp;&nbsp;&nbsp;&nbsp;-&nbsp;Simon Haykin的《神经网络与机器学习》。 事实上，现在常见的很多机器学习算法都发端于神经网络，像SVM，深度学习，CNN等等。这本书详细的介绍了神经网络及其相关算法的所有细节。如果想深入了解的话，可以看一下。只想运用的话，也可以随便翻翻算法的介绍。</span> 
     </div> 
     <div style="background-color:inherit; line-height:20px"> 
      <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">&nbsp;&nbsp;&nbsp;&nbsp;-&nbsp;AIMA，《人工智能：一种现代的方法》。基本上学术界的人们都认为机器学习是人工智能学科的下属分支(另一部分人认为是统计学或者数学的分支)，所以，一本人工智能的书也是学习机器学习可以参考的方面。</span> 
     </div> 
    </div> 
    <hr style="font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> <p style="margin-top:5px; margin-bottom:5px; font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> <span style="font-family:Arial,Helvetica,sans-serif; font-size:12px; background-color:inherit">另外，如果想找这方面的工作，尤其是大企业的工作，主要看你发了哪些论文，开发了哪些系统，而不是看你用XX开源软件多熟练。你懂的，毕竟熟练使用一个软件并不是这个领域的关键问题。</span></p> </td> 
  </tr> 
 </tbody> 
</table>

*  [tornadomeet 机器学习 笔记][tornadomeet _] (2013) -- \[实战笔记\] 学霸的学习笔记，看看小伙伴是怎样一步一步地掌握“机器学习”，好好跟学霸学习机器学习，会有很大的收获。
 *  [Machine Learning Roadmap: Your Self-Study Guide to Machine Learning][Machine Learning Roadmap_ Your Self-Study Guide to Machine Learning] (2014) Jason Brownlee -- \[行家导读\] 虽然是英文版，但非常容易读懂。对Beginner,Novice,Intermediate,Advanced读者都有覆盖。
    
     *  [A Tour of Machine Learning Algorithms][] （2013） 这篇关于机器学习算法分类的文章也非常好
     *  [Best Machine Learning Resources for Getting Started][]（2013） 这片有中文翻译 [机器学习的最佳入门学习资源 @伯乐在线][Link 3] 译者 [programmer\_lin][programmer_lin]
 *  吴恩达在公开课上的讲课笔记网站：*http://www.holehouse.org/mlclass/index.html*
 *  知乎回答精选

<table> 
 <tbody> 
  <tr> 
   <td> <p style="margin-top:5px; margin-bottom:5px; font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> </p> 
    <ul> 
     <li><span style="widows:auto">知乎链接：<u><em>http://www.zhihu.com/question/22928694</em></u>。 &nbsp; &nbsp;&nbsp;以下是最佳回答，觉得值得推荐的书籍：</span></li> 
    </ul> <p></p> 
    <ol style="margin-top:0px; margin-bottom:0px; font-family:微软雅黑; font-size:14px; line-height:21px; widows:auto"> 
     <li style="background-color:inherit">Matrix Cookbook，有免费下载（<a href="http://www.mit.edu/~wingated/stuff_i_use/matrix_cookbook.pdf" style="background-color:inherit" rel="nofollow">http://www.mit.edu/~wingated/stuff_i_use/matrix_cookbook.pdf</a>）<br style="background-color:inherit"> 我特别喜欢的一本的书，听名字就知道是喜闻乐见的深入浅出的良心书...基本上涵盖了所有基础的Matrix和Linear Algebra的知识。可以当做入门教材，也可以当工具书来查阅。</li> 
     <li style="background-color:inherit">Linear Algebra Done Right，第二版（<a href="http://linear.axler.net/" style="background-color:inherit" rel="nofollow">Linear Algebra Done Right</a>），好像里面还有中文版，没看过。可以轻松搜到免费下载。<br style="background-color:inherit"> 我Graduate阶段看的第一本Linear Algebra的书，纠正了很多我以前本科学习时候的错误。</li> 
     <li style="background-color:inherit">Matrix Analysis and Applied Linear Algebra，（<a href="http://www.matrixanalysis.com/" style="background-color:inherit" rel="nofollow">Matrix Analysis &amp; Applied Linear Algebra</a>） 没有免费版<br style="background-color:inherit"> 有一些Vector space的理解以及function analysis的东西。对信号处理里有帮助。<br style="background-color:inherit"> </li> 
    </ol> <p style="margin-top:5px; margin-bottom:5px; font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> 其实Matrix的知识只是工具，我觉得看了以上三本之后，对科研什么的绝对够用了。关键还要看你是做哪方面的应用，需要学习不同的matrix知识。 &nbsp;&nbsp;</p> <p style="margin-top:5px; margin-bottom:5px; font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> <br> </p> <p style="margin-top:5px; margin-bottom:5px; font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> </p> 
    <ul> 
     <li><span style="widows:auto">知乎链接：</span><span style="widows:auto"><u><em>http://www.zhihu.com/question/20691338</em></u>。 &nbsp; &nbsp;&nbsp;最佳回答：</span></li> 
    </ul> <p></p> <p style="margin-top:5px; margin-bottom:5px; font-family:微软雅黑; font-size:14px; widows:auto; line-height:20px"> <span style="color:rgb(34,34,34); font-family:'Helvetica Neue',Helvetica,Arial,sans-serif; font-size:13.3333339691162px; line-height:24.5555572509766px">1. 数学基础</span></p> <p style="padding-top:0px; padding-bottom:0px; margin-top:0px; margin-bottom:0px; color:rgb(34,34,34); font-family:'Helvetica Neue',Helvetica,Arial,sans-serif; font-size:13.3333339691162px; line-height:24.5555572509766px"> 机器学习必要的数学基础主要包括：多元微积分，线性代数</p> 
    <ol style="padding:0px; margin:0.8em 0px 0.8em 2em; list-style-position:outside; color:rgb(34,34,34); font-family:'Helvetica Neue',Helvetica,Arial,sans-serif; font-size:13.3333339691162px; line-height:24.5555572509766px"> 
     <li style="padding:0px; margin:0px 0px 5px"><a href="https://www.coursera.org/course/calcsing" style="" rel="nofollow">Calculus: Single Variable<span style="width:7px; height:7px; display:inline-block; vertical-align:0px; margin-left:4px"></span></a>&nbsp;|&nbsp;<a href="https://www.coursera.org/learn/calculus1" style="" rel="nofollow">Calculus One<span style="width:7px; height:7px; display:inline-block; vertical-align:0px; margin-left:4px"></span></a>&nbsp;（可选）</li> 
     <li style="padding:0px; margin:0px 0px 5px"><a href="http://ocw.mit.edu/courses/mathematics/18-02sc-multivariable-calculus-fall-2010/" style="" rel="nofollow">Multivariable Calculus<span style="width:7px; height:7px; display:inline-block; vertical-align:0px; margin-left:4px"></span></a></li> 
     <li style="padding:0px; margin:0px 0px 5px"><a href="http://ocw.mit.edu/courses/mathematics/18-06-linear-algebra-spring-2010/" style="" rel="nofollow">Linear Algebra<span style="width:7px; height:7px; display:inline-block; vertical-align:0px; margin-left:4px"></span></a></li> 
    </ol> <span style="color:rgb(34,34,34); font-family:'Helvetica Neue',Helvetica,Arial,sans-serif; font-size:13.3333339691162px; line-height:24.5555572509766px">2. 统计基础</span> 
    <ol style="padding:0px; margin:0.8em 0px 0.8em 2em; list-style-position:outside; color:rgb(34,34,34); font-family:'Helvetica Neue',Helvetica,Arial,sans-serif; font-size:13.3333339691162px; line-height:24.5555572509766px"> 
     <li style="padding:0px; margin:0px 0px 5px"><a href="https://www.edx.org/course/introduction-statistics-descriptive-uc-berkeleyx-stat2-1x" style="" rel="nofollow">Introduction to Statistics: Descriptive Statistics<span style="width:7px; height:7px; display:inline-block; vertical-align:0px; margin-left:4px"></span></a></li> 
     <li style="padding:0px; margin:0px 0px 5px"><a href="http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-041-probabilistic-systems-analysis-and-applied-probability-fall-2010" style="" rel="nofollow">Probabilistic Systems Analysis and Applied Probability<span style="width:7px; height:7px; display:inline-block; vertical-align:0px; margin-left:4px"></span></a>&nbsp;|&nbsp;<a href="https://www.coursera.org/course/prob" style="" rel="nofollow">概率<span style="width:7px; height:7px; display:inline-block; vertical-align:0px; margin-left:4px"></span></a>&nbsp;( 可选)</li> 
     <li style="padding:0px; margin:0px 0px 5px"><a href="https://www.edx.org/course/introduction-statistics-inference-uc-berkeleyx-stat2-3x#.U3nU2vmSxhQ" style="" rel="nofollow">Introduction to Statistics: Inference<span style="width:7px; height:7px; display:inline-block; vertical-align:0px; margin-left:4px"></span></a></li> 
    </ol> <span style="color:rgb(34,34,34); font-family:'Helvetica Neue',Helvetica,Arial,sans-serif; font-size:13.3333339691162px; line-height:24.5555572509766px">3. 编程基础</span> 
    <ol style="padding:0px; margin:0.8em 0px 0.8em 2em; list-style-position:outside; color:rgb(34,34,34); font-family:'Helvetica Neue',Helvetica,Arial,sans-serif; font-size:13.3333339691162px; line-height:24.5555572509766px"> 
     <li style="padding:0px; margin:0px 0px 5px"><a href="https://www.coursera.org/course/pythonlearn" style="" rel="nofollow">Programming for Everybody<span style="width:7px; height:7px; display:inline-block; vertical-align:0px; margin-left:4px"></span></a>&nbsp;(Python)</li> 
     <li style="padding:0px; margin:0px 0px 5px"><a href="https://www.google.com.hk/url?sa=t&amp;rct=j&amp;q=&amp;esrc=s&amp;source=web&amp;cd=1&amp;ved=0CB0QFjAA&amp;url=https%3A%2F%2Fwww.datacamp.com%2F&amp;ei=Fa6UVeeAN6O0mAXhrpOgDQ&amp;usg=AFQjCNGbigbocD4QuZU0kvKW8nutUkFBAQ&amp;sig2=XcyIw1Z4iVygtlinr1d29w" style="" rel="nofollow">DataCamp: Learn R with R tutorials and coding challenges<span style="width:7px; height:7px; display:inline-block; vertical-align:0px; margin-left:4px"></span></a>(R)</li> 
     <li style="padding:0px; margin:0px 0px 5px"><a href="https://www.udacity.com/course/intro-to-computer-science--cs101" style="" rel="nofollow">Introduction to Computer Science:Build a Search Engine &amp; a Social Network<span style="width:7px; height:7px; display:inline-block; vertical-align:0px; margin-left:4px"></span></a></li> 
    </ol> <span style="color:rgb(34,34,34); font-family:'Helvetica Neue',Helvetica,Arial,sans-serif; font-size:13.3333339691162px; line-height:24.5555572509766px">4. 机器学习</span> 
    <ol style="padding:0px; margin:0.8em 0px 0.8em 2em; list-style-position:outside; color:rgb(34,34,34); font-family:'Helvetica Neue',Helvetica,Arial,sans-serif; font-size:13.3333339691162px; line-height:24.5555572509766px"> 
     <li style="padding:0px; margin:0px 0px 5px"><a href="https://lagunita.stanford.edu/courses/HumanitiesandScience/StatLearning/Winter2015/about" style="" rel="nofollow">Statistical Learning(R)<span style="width:7px; height:7px; display:inline-block; vertical-align:0px; margin-left:4px"></span></a></li> 
     <li style="padding:0px; margin:0px 0px 5px"><a href="https://www.coursera.org/learn/machine-learning" style="" rel="nofollow">Machine Learning<span style="width:7px; height:7px; display:inline-block; vertical-align:0px; margin-left:4px"></span></a></li> 
     <li style="padding:0px; margin:0px 0px 5px"><a href="https://www.coursera.org/course/ntumlone" style="" rel="nofollow">机器学习基石<span style="width:7px; height:7px; display:inline-block; vertical-align:0px; margin-left:4px"></span></a></li> 
     <li style="padding:0px; margin:0px 0px 5px"><a href="https://www.coursera.org/course/ntumltwo" style="" rel="nofollow">机器学习技法<span style="width:7px; height:7px; display:inline-block; vertical-align:0px; margin-left:4px"></span></a></li> 
    </ol> <p style="padding-top:0px; padding-bottom:0px; margin-top:0px; margin-bottom:0px; color:rgb(34,34,34); font-family:'Helvetica Neue',Helvetica,Arial,sans-serif; font-size:13.3333339691162px; line-height:24.5555572509766px"> 下面是近期的给外行人读的泛数学科普书籍，由浅至深，作用除了感受数学之美之外，更重要的是可以作用每天学习的鸡血，因为这些书都比较好读……</p> <p style="padding-top:0px; padding-bottom:0px; margin-top:0px; margin-bottom:0px; color:rgb(34,34,34); font-family:'Helvetica Neue',Helvetica,Arial,sans-serif; font-size:13.3333339691162px; line-height:24.5555572509766px"> 1.《数学之美》作者：吴军&nbsp;<br> 2.《 Mathematician's Lament | 数学家的叹息》作者：by Paul Lockhart<br> 3.《 Think Stats: Probability and Statistics for Programmers | 统计思维：程序员数学之概率统计 》 作者：Allen B. Downey<br> 4.《 A History of Mathematics | 数学史 》作者：Carl B. Boyer<br> 5.《 Journeys Through Genius | 天才引导的历程：数学中的伟大定理 》作者：William Dunham<br> 6.《 The Mathematical Experience | 数学经验 》作者 Philip J.Davis、Reuben Hersh<br> 7.《 Proofs from the Book | 数学天书中的证明 》作者：Martin Aigner、Günter M. Ziegler<br> 8.《 Proofs and Refutations | 证明与反驳－数学发现的逻辑 》作者：Imre Lakatos</p> </td> 
  </tr> 
 </tbody> 
</table>

[20150718091935940]: /images/20220731/39214f1e3ce04c03afc891802cb87350.png
[http_www.nltk.org_book_ch06.html]: http://www.nltk.org/book/ch06.html
[20150718092108064]: /images/20220731/98a79ccf26c74952ac900cc66bef12cc.png
[http_work.caltech.edu_library_181.html]: http://work.caltech.edu/library/181.html
[20150718092055268]: /images/20220731/bab1109fda1d4b2b8b77dbbb1b0ef7b1.png
[http_n-chandra.blogspot.com_2013_01_picking-machine-learning-algorithm.html]: http://n-chandra.blogspot.com/2013/01/picking-machine-learning-algorithm.html
[20150718092156775]: /images/20220731/f5dba01457634112b6cd0d51c146781e.png
[http_nirvacana.com_thoughts_becoming-a-data-scientist]: http://nirvacana.com/thoughts/becoming-a-data-scientist/
[Link 1]: http://www.guokr.com/post/512037/
[Link 2]: http://www.guokr.com/group/i/0373595356/
[_ourcoders]: http://ourcoders.com/thread/show/2837/
[reyoung]: http://ourcoders.com/user/show/25895/reyoung/
[tornadomeet _]: http://www.cnblogs.com/tornadomeet/tag/%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0/
[Machine Learning Roadmap_ Your Self-Study Guide to Machine Learning]: https://machinelearningmastery.com/machine-learning-roadmap-your-self-study-guide-to-machine-learning/
[A Tour of Machine Learning Algorithms]: http://machinelearningmastery.com/a-tour-of-machine-learning-algorithms/
[Best Machine Learning Resources for Getting Started]: http://machinelearningmastery.com/best-machine-learning-resources-for-getting-started/
[Link 3]: http://blog.jobbole.com/56256/
[programmer_lin]: http://www.jobbole.com/members/linwenhui/