I'm sure you all know about www.ml-class.org - if I were to distillate all lectures to bare minimum it would be to learn about support vector machines (SVM).
Contrary to common believes SVMs are a VERY large class of learners. Linear SVM would be just 1 of 10^10 ways of using SVM. It can be used to do regression, classification, clustering, feature selection, feature weighting, feature engineering, projections, and visualizations. That's why I decided to post few links to the prof. Ng's videos (in the comments section to have clickable URLs).
Model selection http://www.ml-class.org/course/video/embed?video_id=62&s...
Diagnosing bias vs. variance http://www.ml-class.org/course/video/embed?video_id=63&s...
SVM Optimization Objective http://www.ml-class.org/course/video/embed?video_id=72&s...
Large Margin Intuition http://www.ml-class.org/course/video/embed?video_id=73&s...
Mathematics Behind Large Margin Classification (Obligatory!!!) http://www.ml-class.org/course/video/embed?video_id=74&s...
Kernels I http://www.ml-class.org/course/video/embed?video_id=75&s...
Kernels II http://www.ml-class.org/course/video/embed?video_id=76&s...
Using An SVM http://www.ml-class.org/course/video/embed?video_id=77&s...
... yep, you don't need to login.