Monday, June 12, 2006

Beysian Methods

Bayesian for Clustering

HLT 2006 tutorial
http://bayes.hal3.name
http://www.isi.edu/~hdaume/bayes/hlt-slides.pdf

Older HLT 2004 tutorial
https://ssli.ee.washington.edu/~bilmes/bilmes_hlt04_tutorial/bilmes_tutorial.pdf

Non-parametric bayes extends the above with non-parametric prior. This is the hot stuff.
http://www.cs.berkeley.edu/~jordan/nips-tutorial05.ps

More application oriented tutorial
http://www.stanford.edu/~grenager/papers/dp_2005_02_24.ppt
http://www.milab.is.tsukuba.ac.jp/~myama/pdf/topic2006.pdf (This one is in Japanese)

Other presentations
http://l2r.cs.uiuc.edu/~cogcomp/AIML/ARCHIVE/2004-FALL/papers/braz.ppt
http://www.cs.columbia.edu/~risi/talks/chinese.pdf
http://www.cs.huji.ac.il/course/2004/learns/LatentDirichletAlloc.ppt
http://courses.ece.uiuc.edu/ece598/ffl/paper_presentations/NicolasLoeff_LDA.pdf

class lectures
http://www-inst.eecs.berkeley.edu/~cs281a/fa05/lectures/lectures.html

Specific Models

Chinese Restaurant Process & Chinese Restaurant Franchise Process
http://www.cog.brown.edu/~gruffydd/papers/ncrp.pdf
http://www.cs.berkeley.edu/~ywteh/research/npbayes/nips2004a.pdf
http://www.cse.buffalo.edu/faculty/mbeal/papers/bayana05.pdf
http://www.cs.princeton.edu/~blei/papers/TehJordanBealBlei2004.pdf
http://en.wikipedia.org/wiki/Chinese_restaurant_process

Indian Buffet Process
http://www.cog.brown.edu/~gruffydd/papers/ibpnips.pdf
http://www.cog.brown.edu/~gruffydd/papers/ibptr.pdf

Latent Dirichlet Allocation
http://www.jmlr.org/papers/volume3/blei03a/blei03a.pdf
http://www.cs.princeton.edu/~blei/papers/BleiNgJordan2003.pdf
http://chasen.org/~daiti-m/dist/lda/

Pachinco Allocation
http://www.cs.umass.edu/~mccallum/papers/pam-icml06.pdf

===========================
Bayesian for Regression/Classification

Bayes Point Machine
http://research.microsoft.com/~rherb/bayes_point_machines.htm

HLT2006, Bayes Point Machine outperforming Max Margin approach. (Non-projective Dependency Parsing)
http://ssli.ee.washington.edu/people/duh/papers/BPMParsing.pdf

Gaussian Process
http://www.gaussianprocess.org/