发现通过看看别人的slides,作业学习也是不错的方式,有时候直接读paper会过早陷入details,尤其像我这种什么基础都不好的小白,看看slides先大致有个概念然后再看细节或许也不错

关于graph,主要是random walks相关的东西,似乎是jerry做的东西的基础?
http://www.cs.yale.edu/homes/spielman/462/
http://www.cs.yale.edu/homes/spielman/eigs/

learning theory:
rob nowak, 我们学校ece的,跟jerry合作很紧密:
http://www.ece.wisc.edu/~nowak/SLT07.html
avrim blum的:
http://www.machinelearning.com/
http://www.cs.cmu.edu/~avrim/ML09/index.html
侧重点似乎和rob的不一样

peter bartlett:
http://www.cs.berkeley.edu/~bartlett/courses/281b-sp08/
感觉就是带分析的machine learning...

machine learning:
michael jordan,没什么好说的……
http://www.eecs.berkeley.edu/~pliang/cs294-spring08/
Andrew Ng的课程录像:
http://www.youtube.com/results?search_query=machine+learning+stanford+%22machine+learning%22&as=1&and_queries=machine+learning+stanford&exact_query=machine+learning&or_queries=&negative_queries=&geo_name=stanford+ca&geo_latlong=&search_duration=&search_hl=&search_category_type=specific&search_category=27&search_sort=&uploaded=
大概这课是给本科生的……andrew废话有点多@@

fei sha的,主要是下面有个reading list
http://www-rcf.usc.edu/~feisha/htmls/Teaching_CS599_09_Syllabus.html

Gaussian Process:
Carl Edward Rasmussen,hinton大牛的学生,talk很清楚
http://videolectures.net/mlss07_rasmussen_bigp/
一个resource主页,不知道为什么ivm都放里面了:
http://www.gaussianprocess.org/

topics:
hal主持的讨论班,怎么我们学校就没这么有爱的课呢
kernel:
http://apollonius.cs.utah.edu/mediawiki/index.php/MLRG/spring08
multitask:
http://apollonius.cs.utah.edu/mediawiki/index.php/MLRG/summer08
graphical model
http://apollonius.cs.utah.edu/mediawiki/index.php/MLRG/fall08


manifold只找到这个工具的主页 http://www.math.umn.edu/~wittman/mani/

matlab做数值的一个资源,有在线下载numerical computing with matlab:
http://www.mathworks.com/company/aboutus/founders/clevemoler.html