I came across a lecture on Support Vector Machines and in the lecture they converted a maximization

Jaycee Mathis

Jaycee Mathis

Answered question

2022-05-28

I came across a lecture on Support Vector Machines and in the lecture they converted a maximization problem into a minimization problem. I am wondering how it was done...
M a x 1 | | x | |
is converted into
M i n 1 2 x T x
How was this step achieved..? Many thanks in advance !

Answer & Explanation

a2g1g9x

a2g1g9x

Beginner2022-05-29Added 12 answers

I know nothing about VSM, but probably this was a constrained optimization problem where it is known that the solution is bounded away from 0 (unconstrained it seems to make little sense). Recall that
( | | x | | ) 2 =< x , x >= x T x.
Now, if x is bounded away from 0, then maximizing 1 | | x | | is the same as minimizing | | x | | (subject to the constraint I assume is missing; and maximizing the square root of a value leads to the same solution as maximizing the value). The factor 1 2 is just for convenience to make the derivative prettier. After you calculate a solution with the factor, you back the solution w/o factor out from it.

That's my guess; but as said, I know nothing about this field.

Do you have a similar question?

Recalculate according to your conditions!

New Questions in High school geometry

Ask your question.
Get an expert answer.

Let our experts help you. Answer in as fast as 15 minutes.

Didn't find what you were looking for?