We review the idea of the margin of a classifier, and why that may be a good criterion for measuring a classifier's desirability. Then we consider the computational problem of finding the largest margin linear classifier. At this point we look at our toes with embarassment and note that we have only done work applicable to noise-free data. But we cheer up and show how to create a noise resistant classifier, and then a non-linear classifier. We then look under a microscope at the two things SVMs are renowned for---the computational ability to survive projecting data into a trillion dimensions and the statistical ability to survive what at first sight looks like a classic overfitting trap.
Powerpoint Format: The Powerpoint originals of these slides are freely available to anyone who wishes to use them for their own work, or who wishes to teach using them in an academic institution. Please email Andrew Moore at awm@cs.cmu.edu if you would like him to send them to you. The only restriction is that they are not freely available for use as teaching materials in classes or tutorials outside degree-granting academic institutions.
Advertisment: I have recently joined Google, and am starting up the new Google Pittsburgh office on CMU's campus. We are hiring creative computer scientists who love programming, and Machine Learning is one the focus areas of the office. If you might be interested, feel welcome to send me email: awm@google.com .