Machine Learning Ideas
Machine learning is one of the fastest growing branches of computer science with applications in a variety of fields. The purpose of this book is to introduce the reader to the fundamental principles of machine learning and its characteristic algorithmic paradigms. The book contains an extensive set of fundamental theoretical ideas of machine learning and mathematical calculations, thanks to which these ideas become practical algorithms. Following the presentation of the basic foundations of the discipline, a wide range of topics that have not been sufficiently reflected in previous textbooks are considered: computational complexity of learning, concepts of convexity and stability, important algorithms, including stochastic gradient descent, neural networks and structured inference training, as well as very recent theoretical concepts, for example, the RAS-Bayesian approach and compression boundaries. The publication is aimed at senior students studying computer science, technical sciences, mathematics or statistics, and can also be useful to researchers who want to deepen their theoretical knowledge. It is assumed that the reader is familiar with the basics of probability theory, linear algebra, mathematical analysis and the theory of algorithms.
No reviews found
