What are some good online tutorials on the constrained optimization methods used in lecture (Lagrange multipliers for inequality constraints, and quadratic programming)? I want to better understand why an optimal solution to the problem given to the quadratic solver corresponds to a maximum-margin classifier for the original problem. Many thanks!
p.s. Offline (textbook) references are welcome as well.
What are some good online tutorials on the constrained optimization methods used in lecture (Lagrange multipliers for inequality constraints, and quadratic programming)? I want to better understand why an optimal solution to the problem given to the quadratic solver corresponds to a maximum-margin classifier for the original problem. Many thanks!
p.s. Offline (textbook) references are welcome as well.
Maximizing the margin was reduced to a condition on the norm of w, and that was shown to be equivalent to the QP problem in this segment of the lecture:
__________________ Where everyone thinks alike, no one thinks very much