Conventional applications of neural networks usually predict a single value as a function of given inputs. In forecasting, for example, a standard objective is to predict the future value of some entity of interest on the basis of a time series of past measurements or observations. Typical training schemes aim to minimise the sum of squared deviations between predicted and actual values (the 'targets'), by which, ideally, the network learns the conditional mean of the target given the input. If the underlying conditional distribution is Gaus sian or at least unimodal, this may be a satisfactory approach. However, for a multimodal distribution, the conditional mean does not capture the relevant features of the system, and the prediction performance will, in general, be very poor. This calls for a more powerful and sophisticated model, which can learn the whole conditional probability distribution. Chapter 1 demonstrates that even for a deterministic system and 'be nign' Gaussian observational noise, the conditional distribution of a future observation, conditional on a set of past observations, can become strongly skewed and multimodal. In Chapter 2, a general neural network structure for modelling conditional probability densities is derived, and it is shown that a universal approximator for this extended task requires at least two hidden layers. A training scheme is developed from a maximum likelihood approach in Chapter 3, and the performance ofthis method is demonstrated on three stochastic time series in chapters 4 and 5.
"Sinopsis" puede pertenecer a otra edición de este libro.
Conventional applications of neural networks usually predict a single value as a function of given inputs. In forecasting, for example, a standard objective is to predict the future value of some entity of interest on the basis of a time series of past measurements or observations. Typical training schemes aim to minimise the sum of squared deviations between predicted and actual values (the 'targets'), by which, ideally, the network learns the conditional mean of the target given the input. If the underlying conditional distribution is Gaus sian or at least unimodal, this may be a satisfactory approach. However, for a multimodal distribution, the conditional mean does not capture the relevant features of the system, and the prediction performance will, in general, be very poor. This calls for a more powerful and sophisticated model, which can learn the whole conditional probability distribution. Chapter 1 demonstrates that even for a deterministic system and 'be nign' Gaussian observational noise, the conditional distribution of a future observation, conditional on a set of past observations, can become strongly skewed and multimodal. In Chapter 2, a general neural network structure for modelling conditional probability densities is derived, and it is shown that a universal approximator for this extended task requires at least two hidden layers. A training scheme is developed from a maximum likelihood approach in Chapter 3, and the performance ofthis method is demonstrated on three stochastic time series in chapters 4 and 5.
This volume presents a neural network architecture for the prediction of conditional probability densities - which is vital when carrying out universal approximation on variables which are either strongly skewed or multimodal. Two alternative approaches are discussed: the GM network, in which all parameters are adapted in the training scheme, and the GM-RVFL model which draws on the random functional link net approach. Points of particular interest are: - it examines the modification to standard approaches needed for conditional probability prediction; - it provides the first real-world test results for recent theoretical findings about the relationship between generalisation performance of committees and the over-flexibility of their members; This volume will be of interest to all researchers, practitioners and postgraduate / advanced undergraduate students working on applications of neural networks - especially those related to finance and pattern recognition.
"Sobre este título" puede pertenecer a otra edición de este libro.
EUR 11,95 gastos de envío desde Alemania a España
Destinos, gastos y plazos de envíoEUR 4,70 gastos de envío desde Reino Unido a España
Destinos, gastos y plazos de envíoLibrería: Ria Christie Collections, Uxbridge, Reino Unido
Condición: New. In. Nº de ref. del artículo: ria9781852330958_new
Cantidad disponible: Más de 20 disponibles
Librería: moluna, Greven, Alemania
Condición: New. Dieser Artikel ist ein Print on Demand Artikel und wird nach Ihrer Bestellung fuer Sie gedruckt. Provides unique, comprehensive coverage of generalisation and regularisation: Provides the first real-world test results for recent theoretical findings on the generalisation performance of committeesConventional applications of neural networks usually . Nº de ref. del artículo: 4289360
Cantidad disponible: Más de 20 disponibles
Librería: Roland Antiquariat UG haftungsbeschränkt, Weinheim, Alemania
Softcover. XXIII, 275 S. : graph. Darst. ; 24 cm Like new. Unread book. --- Neuwertiger Zustand. Ungelesenes Buch. 9781852330958 Sprache: Deutsch Gewicht in Gramm: 467 Softcover reprint of the original 1st ed. 1999. Nº de ref. del artículo: 200027
Cantidad disponible: 1 disponibles
Librería: AHA-BUCH GmbH, Einbeck, Alemania
Taschenbuch. Condición: Neu. Druck auf Anfrage Neuware - Printed after ordering - Conventional applications of neural networks usually predict a single value as a function of given inputs. In forecasting, for example, a standard objective is to predict the future value of some entity of interest on the basis of a time series of past measurements or observations. Typical training schemes aim to minimise the sum of squared deviations between predicted and actual values (the 'targets'), by which, ideally, the network learns the conditional mean of the target given the input. If the underlying conditional distribution is Gaus sian or at least unimodal, this may be a satisfactory approach. However, for a multimodal distribution, the conditional mean does not capture the relevant features of the system, and the prediction performance will, in general, be very poor. This calls for a more powerful and sophisticated model, which can learn the whole conditional probability distribution. Chapter 1 demonstrates that even for a deterministic system and 'be nign' Gaussian observational noise, the conditional distribution of a future observation, conditional on a set of past observations, can become strongly skewed and multimodal. In Chapter 2, a general neural network structure for modelling conditional probability densities is derived, and it is shown that a universal approximator for this extended task requires at least two hidden layers. A training scheme is developed from a maximum likelihood approach in Chapter 3, and the performance ofthis method is demonstrated on three stochastic time series in chapters 4 and 5. Nº de ref. del artículo: 9781852330958
Cantidad disponible: 1 disponibles
Librería: California Books, Miami, FL, Estados Unidos de America
Condición: New. Nº de ref. del artículo: I-9781852330958
Cantidad disponible: Más de 20 disponibles
Librería: THE SAINT BOOKSTORE, Southport, Reino Unido
Paperback / softback. Condición: New. This item is printed on demand. New copy - Usually dispatched within 5-9 working days 497. Nº de ref. del artículo: C9781852330958
Cantidad disponible: Más de 20 disponibles
Librería: Chiron Media, Wallingford, Reino Unido
Paperback. Condición: New. Nº de ref. del artículo: 6666-IUK-9781852330958
Cantidad disponible: 10 disponibles
Librería: buchversandmimpf2000, Emtmannsberg, BAYE, Alemania
Taschenbuch. Condición: Neu. This item is printed on demand - Print on Demand Titel. Neuware -Conventional applications of neural networks usually predict a single value as a function of given inputs. In forecasting, for example, a standard objective is to predict the future value of some entity of interest on the basis of a time series of past measurements or observations. Typical training schemes aim to minimise the sum of squared deviations between predicted and actual values (the 'targets'), by which, ideally, the network learns the conditional mean of the target given the input. If the underlying conditional distribution is Gaus sian or at least unimodal, this may be a satisfactory approach. However, for a multimodal distribution, the conditional mean does not capture the relevant features of the system, and the prediction performance will, in general, be very poor. This calls for a more powerful and sophisticated model, which can learn the whole conditional probability distribution. Chapter 1 demonstrates that even for a deterministic system and 'be nign' Gaussian observational noise, the conditional distribution of a future observation, conditional on a set of past observations, can become strongly skewed and multimodal. In Chapter 2, a general neural network structure for modelling conditional probability densities is derived, and it is shown that a universal approximator for this extended task requires at least two hidden layers. A training scheme is developed from a maximum likelihood approach in Chapter 3, and the performance ofthis method is demonstrated on three stochastic time series in chapters 4 and 5.Springer Verlag GmbH, Tiergartenstr. 17, 69121 Heidelberg 300 pp. Englisch. Nº de ref. del artículo: 9781852330958
Cantidad disponible: 1 disponibles
Librería: Books Puddle, New York, NY, Estados Unidos de America
Condición: New. pp. 302. Nº de ref. del artículo: 263158425
Cantidad disponible: 4 disponibles
Librería: Revaluation Books, Exeter, Reino Unido
Paperback. Condición: Brand New. 275 pages. 9.50x6.25x0.75 inches. In Stock. Nº de ref. del artículo: x-1852330953
Cantidad disponible: 2 disponibles