Skip to content

Booksellers & Trade Customers: Sign up for online bulk buying at trade.atlanticbooks.com for wholesale discounts

Booksellers: Create Account on our B2B Portal for wholesale discounts

Neural Networks for Conditional Probability Estimation: Forecasting Beyond Point Predictions

by Dirk Husmeier
Save 35% Save 35%
Current price ₹3,639.00
Original price ₹5,597.00
Original price ₹5,597.00
Original price ₹5,597.00
(-35%)
₹3,639.00
Current price ₹3,639.00

Imported Edition - Ships in 12-14 Days

Free Shipping in India on orders above Rs. 500

Request Bulk Quantity Quote
+91
Book cover type: Paperback
  • ISBN13: 9781852330958
  • Binding: Paperback
  • Subject: N/A
  • Publisher: Springer
  • Publisher Imprint: Springer
  • Publication Date:
  • Pages: 275
  • Original Price: EUR 49.99
  • Language: English
  • Edition: Softcover Repri
  • Item Weight: 468 grams
  • BISAC Subject(s): Data Science / Neural Networks and Artificial Intelligence / Computer Vision & Pattern Recognition

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.

Trusted for over 49 years

Family Owned Company

Secure Payment

All Major Credit Cards/Debit Cards/UPI & More Accepted

New & Authentic Products

India's Largest Distributor

Need Support?

Whatsapp Us