Permalink
Switch branches/tags
Nothing to show
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
75 lines (42 sloc) 3.54 KB

Bayesian network 与python概率编程实战入门

contributors: @西瓜大丸子汤 @王威廉 @不确定的世界2012 @Rebecca1020

1. Bayesian network 入门讲义幻灯片

http://www.cs.cmu.edu/~epxing/Class/10708/lectures/lecture2-BNrepresentation.pdf Directed Graphical Models: Bayesian Networks

  • 王威廉 推荐

http://www.ee.columbia.edu/~vittorio/Lecture12.pdf Inference and Learning in Bayesian Networks

http://courses.cs.washington.edu/courses/cse515/09sp/slides/bnets.pdf Bayesian networks

2. 基于python的实战入门

Bayesian Methods for Hackers 6000+ star book on github

Frequentists and Bayesians series four blogs

PyMC tutorial pretty short


补充相关材料

基于R的实战入门

http://site.douban.com/182577/widget/notes/12817482/note/273585095/ 贝叶斯网的R实现( Bayesian network in R)

相关进阶

http://bayes.cs.ucla.edu/BOOK-2K/index.html Causality: Models, Reasoning, and Inference 

http://www.biostat.jhsph.edu/~cfrangak/papers/preffects.pdf Principal Stratification in Causal Inference - Biostatistics (2002)

  • Don Rubin
  • Rebecca1020 推荐

http://www.cs.cmu.edu/~epxing/Class/10708/lecture.html Probabilistic Graphical Models by Eric Xing(CMU)

  • 王威廉 推荐

http://web4.cs.ucl.ac.uk/staff/D.Barber/pmwiki/pmwiki.php?n=Brml.HomePage Bayesian Reasoning and Machine Learning by David Barber

  • @诸神善待民科组 推荐 (比 Koller 的 PGM 好读,好处是图多)

相关微博

@王威廉 :CMU机器学习系Eric Xing老师的Probabilistic Graphic Model 已经开了10个年头了, 这学期貌似是第一次把视频放在网上:http://t.cn/zTh9OqO 目前这学期的课程刚开始。 1月23日15:21 http://weibo.com/1657470871/AtrlldqAU

@西瓜大丸子汤 :在推荐一本我最近正在看的书Probabilistic Programming and Bayesian Methods for Hackers 贝叶斯方法实战,用Python来解释各种概率推理方法,有代码有真相。基于PyMC 包,解剖了MCMC ,大数定律,金融分析等概念与应用。Github上已经有5000颗星。 7月8日20:06 http://weibo.com/1932835417/BcKj0k0Wx

@不确定的世界2012 :【贝叶斯网的R实现( Bayesian network in R)(一)gRain(1)】#本文主要介绍运用贝叶斯网的一些R语言工具。 贝叶斯网,又称信念网络或概率有向无环图模型(Bayesian network,belief network,probabilistic directed acyclic graphical m... http://t.cn/zToro0U 2013-7-2 19:52 http://weibo.com/1768506843/zEfzDsln9

@Rebecca1020 :因果推断在USA分两大学派:因果推断本质统计做不了的,但为了能得到inference,必须要加入假设。不同假设就产生了两大不同的学派。西边以berkeley为主,Jordan他们搞的是bayesian network,用有向图来代表之间因果关系。东边Rubin在03年提出Principal stratification,以此为主要假设来进行统计推断。 2013-4-11 02:44 http://weibo.com/1669820502/zrFNJv8DI

张天雷 提供中文介绍《概率编程语言与贝叶斯方法实践》 //@小猴机器人: 来,给个中文介绍哈, http://t.cn/RPwbEPz http://www.weibo.com/5220650532/BmkyPihT4