Springer Learning and Generalisation - Neural Networks Book

Explore the formal mathematical theory behind machine learning and neural networks with this comprehensive text from Springer. This book addresses fundamental questions regarding how machines learn concepts from examples and how neural networks predict outcomes for previously unseen inputs. It provides a rigorous framework for understanding the relationship between training data and predictive accuracy. This second edition expands on previous concepts to include vital modern topics such as support vector machines and fat-shattering dimension. It is designed for those looking to understand the dynamical behavior of nonlinear control systems by observing input-output behavior over finite time periods. Whether you are studying the requirements for prediction accuracy or the mechanics of neural network training, this volume offers the mathematical depth needed to master these complex subjects.