Evolution of Deep Learning from Turing machine to Deep Learning next generation.

Despite the increasing of research papers, methodological developments, and applications of Deep Learning algorithms, a paper on the history of these models is still missing. In this study, it is provided a biography of Deep Learning, starting from its origin to this (very) moment considering the most relevant results. The history of Deep Learning is particularly interesting; indeed, as probably never before, it was born out of the interaction of different expertise, and now we are often in touch with technologies based on these algorithms. Indeed, the first definition of neuron has been possible only for the synergy between a psychologist/neuro-anatomist, McCulloch, and a mathematician, Pitts. Together they laid the foundations for what we now call Deep Learning. In this paper, the history with the most significant intuitions is shown, as, to our knowledge, it has never been done in previous literature. This work aims at covering this lack, presenting a chronological history of the evolution from the first neuron to today's sophisticated Evolutionary Computing, and providing the most relevant references for each addressed issue.