Machine Learning: A Constraint-Based Approach - Comprehensive Guide for AI & Data Science | Perfect for Researchers, Developers & Students

Those who are lucky enough to know Marco Gori personally are well aware of his capacity to talk about virtually any topic in science (and broad surroundings) competently and pleasantly while, above all, always doing so from a non-trivial perspective. Either in front of his students' audience in a classroom or at conferences, as well as in front of a glass of wine in a relaxed, informal environment, one can enjoy his clear, yet colorful and descriptive speech. For those who do not know him (and are interested in machine learning), this book represents a perfect chance to understand what I mean.Machine learning is an extensive research and application field, whose goals can be achieved by methods which are the outcome of the work of different research groups. This usually causes a perceivable bias, induced by the author's belonging to one of such communities, when machine learning is dealt with in the scientific literature. By viewing machine learning as a set of methods for solving constraint-satisfaction problems, Marco Gori offers a clear unified, broad and unbiased overview of the field.It is a very comprehensive and well-organized textbook, with an excellent set of non-trivial exercises supporting each chapter, even if possibly not precisely a primer on machine learning. More relevantly, it is a must-read for the researcher or practitioner in machine learning, at any skill level, to whom the original perspective from which the author treats machine learning topics will unveil a world of connections between different methods which are usually hard to detect in more traditional textbooks.