Cookies on oxfam

We use cookies to ensure that you have the best experience on our website. If you continue browsing, we’ll assume that you are happy to receive all our cookies. You can change your cookie settings at any time. Find out more Close

  • Grows vegetables
  • Fills classrooms
  • Drills wells
  • Empowers women
  • Fights poverty

Second-order Methods for Neural Networks / Adrian J. Shepherd

£4.99 Out Of Stock

Product description

Subtitle: Fast and Reliable Training Methods for Multi-Layer Perceptrons. This book is in excellent condition - almost as new inside and out.

This book is about training methods - in particular, fast second-order training methods - for multi-layer perceptrons (MLPs). MLPs (also known as feed-forward neural networks) are the most widely-used class of neural network. Over the past decade MLPs have achieved increasing popularity among scientists, engineers and other professionals as tools for tackling a wide variety of information processing tasks. In common with all neural networks, MLPs are trained (rather than programmed) to carryout the chosen information processing function. Unfortunately, the 'traditional' method for training MLPs- the well-known backpropagation method - is notoriously slow and unreliable when applied to many practical tasks. The development of fast and reliable training algorithms for MLPs is one of the most important areas of research within the entire field of neural computing. The main purpose of this book is to bring to a wider audience a range of alternative methods for training MLPs, methods which have proved orders of magnitude faster than backpropagation when applied to many training tasks. The book also addresses the well-known 'local minima' problem, and explains ways in which fast training methods can be combined with strategies for avoiding (or escaping from) local minima. All the methods described in this book have a strong theoretical foundation, drawing on such diverse mathematical fields as classical optimisation theory, homotopic theory and stochastic approximation theory.

Item details

Author(s):
Adrian J. Shepherd
Condition:
Used: very good
Dimensions:
15 x 23 cm
Edition:
1st
Format:
Paperback
ISBN-10:
3540761004
ISBN-13:
9783540761006
Number of pages:
145
Publisher:
Springer (1997)
Title:
Second-order Methods for Neural Networks

Sold Out

We're sorry to say this item has now sold out.

There are over 100,000 more items in stock. Start browsing with our suggestions below.

About this item

Subtitle: Fast and Reliable Training Methods for Multi-Layer Perceptrons. This book is in excellent condition - almost as new inside and out.

This book is about training methods - in particular, fast second-order training methods - for multi-layer perceptrons (MLPs). MLPs (also known as feed-forward neural networks) are the most widely-used class of neural network. Over the past decade MLPs have achieved increasing popularity among scientists, engineers and other professionals as tools for tackling a wide variety of information processing tasks. In common with all neural networks, MLPs are trained (rather than programmed) to carryout the chosen information processing function. Unfortunately, the 'traditional' method for training MLPs- the well-known backpropagation method - is notoriously slow and unreliable when applied to many practical tasks. The development of fast and reliable training algorithms for MLPs is one of the most important areas of research within the entire field of neural computing. The main purpose of this book is to bring to a wider audience a range of alternative methods for training MLPs, methods which have proved orders of magnitude faster than backpropagation when applied to many training tasks. The book also addresses the well-known 'local minima' problem, and explains ways in which fast training methods can be combined with strategies for avoiding (or escaping from) local minima. All the methods described in this book have a strong theoretical foundation, drawing on such diverse mathematical fields as classical optimisation theory, homotopic theory and stochastic approximation theory.

Author(s):
Adrian J. Shepherd
Condition:
Used: very good
Dimensions:
15 x 23 cm
Edition:
1st
Format:
Paperback
ISBN-10:
3540761004
ISBN-13:
9783540761006
Number of pages:
145
Publisher:
Springer (1997)
Title:
Second-order Methods for Neural Networks

Delivery & returns

This item will be dispatched to UK addresses via second class post within 2 working days of receipt of your order. Standard UK delivery is £3.95 per order, so you're only charged once no matter how many items you have in your basket. Any additional courier charges will be applied at checkout as they vary depending on delivery address.

This item will be dispatched to UK addresses via second class post within 2 working days of receipt of your order. Standard UK delivery is currently free, no matter how many items you have in your basket. Any additional courier charges will be applied at checkout as they vary depending on delivery address.

You can find out more about delivery and returns in our help section.

We offer a no quibble returns policy as follows:

Wedding dresses: 14 days

Overseas returns: 31 days

Everything else: 21 days


This item is also available for international delivery by airmail, carrying a mandatory delivery charge of:

Europe: £4.70

Outside Europe: £7.00

Volunteer listed

Wonder how this unique item ended up online?

Most of the second-hand items you see online have been donated, by supporters like you, to our high street stores. Each item is then priced, photographed and listed on this site by our amazing team of volunteers from across the country.

After you have bought your item, our team of volunteers package and dispatch it from the Shop straight to you or your chosen recipient.

All profits from the sales of our goods go towards funding Oxfam's work around the world. We rely on your donations to sell online so please keep the cycle of goodness going!

To find out more about volunteering with Oxfam, please visit our how to volunteer page.

Oxfam Books & Music Beeston

The Beeston books and music shop is situated near to Nottingham University, close to the city of Nottingham. There is a train station nearby. We offer a wider selection of books than most shops, from modern fiction and leisure interest to more specialist subjects, academic books, and rare or collectable items. There is a wide variety of music, including sheet music and a good selection of high quality classical, rock and pop vinyl. The volunteer team is a diverse mix from the local community, and University. We are currently recruiting for a quality control volunteer. We always welcome good quality, saleable donations which can be delivered to our rear entrance.

For our latest book promotions and events please check our Facebook page and Wordpress site.

View Shop