Lecture notes at theory.stanford.edu/~tim/w16/l/l7.pdf Topics: Introduction to linear programming. Geometric intuition. Applications: maximum and minimum-cost flow; linear regression; learning a linear classifier, with extensions to minimizing hinge loss and augmented feature sets.
The same problem "fitting a line", I have seen two different courses explain it, one is linear algebra, the other is about ML; they all explain this issue pretty well from different angles, but I still have some doubts... Thankfully, they are now cleared by Lecture 7! Thank you, Professor!
Tim, I watch more of your lectures than I do from my own professors. I wouldn't have survived this quarter without you. Thanks!
Lecture notes at theory.stanford.edu/~tim/w16/l/l7.pdf
Topics: Introduction to linear programming. Geometric intuition. Applications: maximum and minimum-cost flow; linear regression; learning a linear classifier, with extensions to minimizing hinge loss and augmented feature sets.
The same problem "fitting a line", I have seen two different courses explain it, one is linear algebra, the other is about ML; they all explain this issue pretty well from different angles, but I still have some doubts... Thankfully, they are now cleared by Lecture 7! Thank you, Professor!
Best teacher ever.
Greattttttt Video ! Thanks for posting the notes !
@01:03:00 regd h(pi)>0, instead of using a δ, can we not simply replace by -h(pi) ≤ 0?