Hinge loss¶
Loss of the form:
\[
L_{\text{hinge}} (y, \eta) = \max(0, 1 - y \eta) = (1 - y\eta)_{+}
\]
\(\eta = f(x)\) is our confidence in choosing label \(y=1\). (Without any probabilistic semantics)
The function looks like a door hinge:
Loss of the form:
\(\eta = f(x)\) is our confidence in choosing label \(y=1\). (Without any probabilistic semantics)
The function looks like a door hinge: