A task-oriented approach to Machine Learning allows us to divide it into three categories: supervised learning (data are labelled according to some criteria), unsupervised learning (no labels exist for the data), and reinforcement learning (no data is directly available, only...