Can SVM hyperplane always separate data?

89 Views Asked by At

According to this post SVM problem is a convex problem with convex constraints. Now what I am really struggling to understand if SVM is a convex problem, can we draw separating hyperplane for every classification problem? Am I thinking it wrongly?

1

There are 1 best solutions below

2
On

By convex optimization problem, it means the objective function is convex and the feasible set is convex.

It doesn't imply that every problem is linearly separable. It is possible for two distinct data having the same feature $x$, and they belong to different classes.