Machine Learning has been at the core of Artificial Intelligence since its inception. Many promises have been held, if one is to consider that Google is a living demonstration of AI. This paper presents a historical perspective on Machine Learning, describing how the emphasis was gradually shifted from logical to statis-tical induction, from induction to optimization, from the search of hypotheses to the search of representa-tions. The paper concludes with a discussion about the new frontier of Machine Learning.