Close
Enter your search into one or more of the boxes below:
You can refine your search by selecting from any of the options below:
Search
Neural Networks for Conditional Probability Estimation: Forecasting Beyond Point Predictions

Neural Networks for Conditional Probability Estimation: Forecasting Beyond Point Predictions (Paperback)

£64.99
Printed to order. Despatched in 2-3 weeks.
Email me when back in stock

Synopsis

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.

Computing & ITComputer scienceArtificial intelligenceNeural networks & fuzzy systemsPhilosophy, Psychology & Social SciencesSocial issues & processesSocial forecasting, future studies Publisher: Springer London Ltd Publication Date: 22/02/1999 ISBN-13: 9781852330958  Details: Type: Paperback Format: Books
Availability: Printed to order. Despatched in 2-3 weeks.  

More books by Dirk Husmeier

Leave Review

Delivery

Delivery Options

All delivery times quoted are the average, and cannot be guaranteed. These should be added to the availability message time, to determine when the goods will arrive. During checkout we will give you a cumulative estimated date for delivery.

Location 1st Book Each additional book Average Delivery Time
UK Standard Delivery FREE FREE 3-5 Days
UK First Class £4.50 £1.00 1-2 Days
UK Courier £7.00 £1.00 1-2 Days
Western Europe** Courier £17.00 £3.00 2-3 Days
Western Europe** Airmail £5.00 £1.50 4-14 Days
USA / Canada Courier £20.00 £3.00 2-4 Days
USA / Canada Airmail £7.00 £3.00 4-14 Days
Rest of World Courier £22.50 £3.00 3-6 Days
Rest of World Airmail £8.00 £3.00 7-21 Days

** Includes Austria, Belgium, Denmark, France, Germany, Greece, Iceland, Irish Republic, Italy, Luxembourg, Netherlands, Portugal, Spain, Sweden and Switzerland.

Delivery Help & FAQs

Returns Information

If you are not completely satisfied with your purchase*, you may return it to us in its original condition with in 30 days of receiving your delivery or collection notification email for a refund. Except for damaged items or delivery issues the cost of return postage is borne by the buyer. Your statutory rights are not affected.

* For Exclusions and terms on damaged or delivery issues see Returns Help & FAQs

You might also like

Convolutional Neural Networks in...
(Paperback)
Baoxin Li
 
 
£60.00
 
Deep Learning and Convolutional ...
(Hardback)
Le Lu; Yefeng Zheng
 
 
£88.00
 
Neural Network Driven Artificial...
(Hardback)
Bahman Zohuri; Masoud Moghaddam
 
 
£219.50
 
Large-Scale Machine Learning in the...
(Hardback)
Ashok N. Srivastava; Ramakrishna...
 
 
£100.00
 
Latest Blog
#FoylesFive: Swap Snowflakes for Stardust
17/11/2017

Swap snowflakes for stardust this Christmas with some recommends that are out of this world, from Jamie-Lee from our Birmingham branch.

Read an extract of The Book of the Year
15/11/2017

Behold The Book of the Year, by the “No Such Thing As A Fish” podcasters a.k.a the QI Elves. The book is a surreal, silly and wonderful compendium of irreverent and topical facts that show 2017 in a unique light.

Japan: the Perfect Setting for a Crime Novel
14/11/2017

Nicolas explains why Japan is the perfect backdrop to a crime novel.

View all Blog Entries
Twitter
Show/Hide Tweets
© W&G Foyle Ltd