18.4 Suppose you are running a learning experiment on a new algorithm.
You have a data set consisting of 25 examples of each two classes. You
plan to use leave-one-out cross validation. As a a baseline, you run
your experimental setup on a simple majority classifier. You expect
the majority classifier to score about 50% on leave-one-out
cross-validation, but to your surprise, it scores 0. Can you explain
why?