Neural Networks and Sea Time Series. Reconstruction and extreme-event analysis

摘要

Nowadays, neural networks (NN) are widely applied to several classification, prediction and approximation tasks. The book under consideration touches one of the less popular application areas, which is analysis, reconstruction and prediction of various marine time series.rnThe book is composed of eleven chapters, which can functionally be divided into four parts. The first two chapters provide an introduction to the subject. Chapters 3-6 cover mathematical foundations of marine time series description and discuss theoretically the use of NN and statistical methods in approximation and prediction of such series. The next part, composed of the four following chapters, describes several experiments carried out using NN and statistical models. The last chapter provides conclusions and a brief description of open problems encountered by the authors while working on the subject.