I took the original Stanford AI course and got an immense amount out of it. I got a poor score since I did not have a great deal of time to work out all the problems.. But what I really enjoyed about the course was getting a grasp of the new techiques, way beyond the conventional statitistcs..

I'm looking for a way to get to the next level. The course really did not provide a good reference for reviewing the subject. The big book by the course authors (Artificial Intelligence: A Modern Approach by Norvig and Russell) (which I have) is too complex for me.

I try to do all my work in Python and I am making some headway in getting the tools, Sci-py, scikit learn, NLTK. Pandas (am currently reading Data Analysis with Python, by Mckinney, excellent). and more.. But I'm in need of references to help me leap from the conceptual level to actual implementation , including proper tool selection and problem selection and definition, that is, sort of a cook book approach.

This question may be too open ended,, but it expresses my dilemma. The whole space is quite open ended, and I'm looking for references to help me navigate through it.

What references (accessible) might you suggest?

  • $\begingroup$ What did you mean by"proble selection"? I wasn't sure how to fix that. I did edit other things to correct spelling and remove slang as well as making some points clearer/ $\endgroup$ – Michael Chernick Sep 30 '12 at 16:48
  • $\begingroup$ Also maybe if you tell us what the course text was, we can have a better idea as to what might be too complex or inaccessible to you. $\endgroup$ – Michael Chernick Sep 30 '12 at 16:50
  • $\begingroup$ Thanks for pointing out my lousy typos! Book title added! $\endgroup$ – dartdog Sep 30 '12 at 22:06
  • $\begingroup$ Coursera Probabilistic Graphical Models class is going on now. $\endgroup$ – user14516 Oct 1 '12 at 1:35
  • $\begingroup$ There are several courses available, I just don't have the time and for the time being think I need more reading before heading back for courses.. I just learn better that way. $\endgroup$ – dartdog Oct 1 '12 at 2:52

I wanted to refer you to the Statistical Learning book but I wanted to be sure that it wasn't your text. It was written by Stanford Statistics faculty Trevor Hastie, Jerome Friedman and Rob Tibshirani. Here is a link to it on amazon.

  • 3
    $\begingroup$ Also can be downloaded at Tibshirani's homepage www-stat.stanford.edu/~tibs/ElemStatLearn $\endgroup$ – ziyuang Sep 30 '12 at 23:56
  • $\begingroup$ I had downloaded it but not yet read, with your recommendation will move up the list! Thanks for the pointer.. There is so much out there it is hard to even determine what to do next, without guidance. Thank you! $\endgroup$ – dartdog Oct 1 '12 at 0:36
  • $\begingroup$ FYI I'd possibly mark this as answered but want to see if others have ideas, as well want to at least skim the book you mentioned. So I'll keep checking! $\endgroup$ – dartdog Oct 1 '12 at 2:53
  • $\begingroup$ You could still give me an upvote even if you decide to check someone else's suggestion. $\endgroup$ – Michael Chernick Oct 1 '12 at 3:09
  • $\begingroup$ That book really is wonderful, read it start to end. $\endgroup$ – bdeonovic Apr 23 '14 at 23:45

I'm finding that the machine learning course from Coursera/Andrew Ng/Stanford is really great. It has a balance of statistics, AI, linear algebra, programming. The motivations/intuitions that are given in the video lectures get me pumped to solve some problems.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.