0
$\begingroup$

I have trouble solving the problem 3.18 from "pattern recognition" by "Sergios Theodoridis, ‎Konstantinos Koutroumbas" the problem is :

Show that for the case of two linearly separable classes the hyperplane obtained as the SVM solution is the same as that bisecting the segment joining two closest points between the convex hulls of the classes.

thank you all

$\endgroup$
0
$\begingroup$

A paper that shows the relationship between SVM and convex hull is

www.robots.ox.ac.uk/~cvrg/bennett00duality.pdf

I don't know if the proof you ask is there, but the picture of separable data and convex hull is there - the intuition becomes clear after reading the beginning of the paper.

$\endgroup$

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.