In this work formulated relevance, set out an analytical review of existing approaches to the research recurrent neural networks (RNN) and defined precondition appearance a new direction in the field neuroinformatics – reservoir computing. Shows generalized classification neural network (NN) and briefly described main types dynamics and modes RNN. Described topology, structure and features of the model NN with different nonlinear functions and with possible areas of progress. Characterized and systematized wellknown learning methods RNN and conducted their classification by categories. Determined the place RNN with unsteady dynamics of other classes RNN. Deals with the main parameters and terminology, which used to describe models RNN. Briefly described practical implementation recurrent neural networks in different areas natural sciences and humanities, and outlines and systematized main deficiencies and the advantages of using different RNN. The systematization of known recurrent neural networks and methods of their study is performed and on this basis the generalized classification of neural networks was proposed.
We consider the problem of visualisation of high dimensional multivariate time series. A data analyst in creating a two dimensional projection of such a time series might hope to gain some intuition into the structure of the original high dimensional data set. We review a method for visualising time series data using an extension of Echo State Networks (ESNs).The method uses the multidimensional scaling criterion in order to create a visualisation of the time series after its representation in the reservoir of the ESN. We illustrate the method with two dimensional maps of a financial time series. The method is then compared with a mapping which uses a fixed latent space and a novel objective function.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.