- In the single-neuron discussion at the start of this section, I argued that the cross-entropy is small if (z)y for all training inputs. The argument relied on y being equal to either 0 or 1 . This is usually true in classification problems, but for other problems (e.g., regression problems) y can sometimes take values intermediate between 0 and 1 . Show that the cross-entropy is still minimized when (z)=y for all training inputs. When this is the case the crossentropy has the value: C=n1x[yln(y)+(1y)ln(1y)] The quantity [yln(y)+(1y)ln(1y)] is sometimes known as the binary entropy