Mathematical foundations of machine learning and prediction

The book is intended for the initial acquaintance with the mathematical foundations of the modern theory of machine learning (Machine Learning) and the theory of games with predictions. The first part sets the foundations of the statistical theory of machine learning, the tasks of classification and regression with support vectors, the theory of generalization and the algorithms for constructing separating hyperprafs are considered. In the second and third parts, the tasks of adaptive forecasting in the non-chronic theoretical-game and comparative productions are considered: predictions using expert strategies (Prediction with Expert Advice) and prediction games.
For students and graduate students specializing in the field of machine learning and artificial intelligence.
2nd edition, corrected and supplemented