What loss function is specifically used by Support Vector Machines to maximize the margin between classes?
Show answer
Hinge loss
Hinge loss is defined as max(0, 1 - y*f(x)), penalizing predictions that are on the wrong side of the margin or the decision boundary. This piecewise linear function is what gives SVMs their sparse solution property, as only support vectors contribute to the loss.
Also asked as:
- Which loss function, defined as max(0, 1 - y*f(x)), is used by Support Vector Machines to penalize predictions on the wrong side of the margin?