One of the most important challenges of ICT engineering is the development of suitable algorithms for automatic complex systems modelling. In fact, the most interesting and useful applications of modern engineering deal with multi-disciplinary topics, closely related to complex systems. It is not by chance, since complex systems are everywhere in nature, as well as in most artificial systems designed and built by humankind. Complex systems are by far more frequent than “simple” ones, which are the true outliers in our world. How and to what extent can we predict the behavior of a complex system? The course aims to introduce Machine Learning and Granular Computing techniques to find regularities in data coming as samples of directed or undirected processes to be modeled, and to discover at the same time the best semantic level where these regularities take place. Effective and efficient implementation of such algorithms demand both the use of massive manycore processing systems, as well the acceleration in hardware of custom dissimilarity measure computations, relying on both GPUs and FPGAs.