有几本书是值得一读的（都可以下到）：
［1］ (N) Neapolitan, R. E. Learning Bayesian Networks. Pearson Prentice
Hall, 2004
［2］ (RN) Russell, S. and Norvig, P. Artificial Intelligence: A Modern
Approach, 2nd ed. Prentice Hall, 2003
［3］(KN) Korb, K. B. and Nicholson, A. E. Bayesian Artificial
Intelligence. Chapman and Hall/CRC, 2004.
我觉得（KN）讲得比较详细。
BN的无痛入门：
Charniak, Eugene “Bayesian Networks without Tears”, AI Magazine,
12(4), Winter 91, 50-63
会议和Murphy的主页：
You will also find valuable tutorials, tools, publications on
Bayesian networks and related technologies at the following
websites:
Conference in Uncertainty in Artificial Intelligence (UAI)
www.auai.org
American Association for Artificial Intelligence Conference (AAAI)
www.aaai.org
International Joint Conference on Artificial Intelligence (IJCAI)
www.ijcai.org
Neural Information Processing Systems Conference (NIPS)
www.nips.cc
Kevin Murphy’s tutorial on Graphical models and Bayesian networks
http://www.cs.ubc.ca/~murphyk/Bayes/bnintro.html
两个著名的研究BN的工具软件：
Genie 2.0 by Decision Systems Laboratory, University of Pittsburgh
http://genie.sis.pitt.edu/
An excellent, free probabilistic networks reasoning program
Netica 4.08 by Norsys Software Corp.
Free versions of the main application and the APIs are available
＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞＞
＞＞＞
以下是介绍各个分支的一些情况：
＠＠＠＠＠＠＠＠＠＠＠＠＠＠＠＠＠
Introduction to Bayesian Network:
＠＠＠＠＠＠＠＠＠＠＠＠＠＠＠＠＠
人物：
Judea Pearl as the 2008 Benjamin Franklin Medal in
Computers and Cognitive Science
http://www.fi.edu/franklinawards/08/laureate_bf_computerccs-
pearl.html
Often regarded as the “Father of Bayesian Networks”
Other pioneers in the field:
Lauritzen, Steffen L.
Spiegelhalter, David J.
Neapolitan, Richard E.
Jensen, Finn V.
... and many others
Homework 1
＊＊＊＊＊＊＊＊
Readings:
1.
Charniak, Eugene “Bayesian Networks without Tears”, AI
Magazine, 12(4), Winter 91, 50-63
N: 1.1-1.2, 1.3 (Italic sections are optional readings)
Download and install:
2.
Genie 2.0 by Decision Systems Laboratory, University of
Pittsburgh http://genie.sis.pitt.edu/
An excellent, free probabilistic networks reasoning program
Netica 4.08 by Norsys Software Corp.
Free versions of the main application and the APIs are available
＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜
＠＠＠＠＠＠＠＠＠＠＠＠＠＠＠＠
Bayesian Network Representation
＠＠＠＠＠＠＠＠＠＠＠＠＠＠＠＠
Readings:
N: 1.4, 2.1 (2.1.2), 2.2, 2.3 (2.3.1), 2.4, 2.5
RN: 14.1-14.2
KN: Chapter 2
R. D. Shachter. “Bayes-ball: The rational pastime.” In
Proceedings of the 14th Conference on Uncertainty in
Artificial Intelligence, pages 480-487, 1998.
＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜
＠＠＠＠＠＠＠＠＠＠＠
Exactly Inference
＠＠＠＠＠＠＠＠＠＠＠
N: Chapter 3
KN: Chapter 3
RN: 14.4
JN: Chapter 4
Guo, H. and W. Hsu (2002). A Survey of Algorithms for Real-
Time Bayesian Network Inference. In Proceedings of the Joint
AAAI-02/KDD-02/UAI-02 Workshop on Real-Time Decision
Support and Diagnosis Systems, Edmonton, Alberta, Canada
一定不能少了这伙计的： Pearl, BN的始祖之一。Judea Pearl as the 2008 Benjamin F
ranklin Medal in
Computers and Cognitive Science http://www.fi.edu/franklinawards/08/laureate_b
f_computerccs-
pearl.html
＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜＜
＠＠＠＠＠＠＠＠＠＠＠
Approximate Inference
＠＠＠＠＠＠＠＠＠＠＠
References
RN: 14.5
KN: 3.6
KF: 8.3, 11
Murphy, K. An introduction to graphical models. May 2001. Unpublished
manuscript, available at: http://www.cs.ubc.ca/~murphyk/Papers/intro_gm.pdf
Murphy, K. Software Packages for Graphical Models / Bayesian Networks. Last
updated 28 July 2008. http://www.cs.ubc.ca/~murphyk/Software/bnsoft.html
Jordan, M.I., Ghahramani, Z., Jaakkola, T.S., and Saul, L.K. (1999) An Introd
uction
to Variational Methods for Graphical Models. Machine Learning, Vol. 37, No. 2,
pp.
183-233.
K. Murphy, Y.Weiss, and M. Jordan. Loopy belief propagation for approximate
inference: An empirical study. In UAI'99, pages 467-475, 1999.
Yedidia, J.S.; Freeman, W.T.; Weiss, Y., "Generalized Belief Propagation",
Advances in Neural Information Processing Systems (NIPS), Vol 13, pps 689-695,
December 2000
Jordan和Bishop的材料（Approximate Inference)：
[PRML] Pattern Recognition and Machine Learning 第10章Approximate Inference 主
要讲了Variational Method; 这个方法现在和belief propagation 是approximate infer
ence 中很HOT的两种方法。
而variational method（变分法） 更是用途广泛：）
Jordan 派的Graphical Model这个自然就不用再说了。
待续：）
--
Vi Veri Veniversum Vivus Vici
I by the power of Truth, while living, have conquered the Universe。
以真理的力量 我在死前征服全宇宙
－－－－浮士德