Introduction to Neural Networks with Java, 1st Edition

Book Name: Introduction to Neural Networks with Java, 1st Edition ISBN: 097732060X Author: Jeff Heaton Pages: 380 Status: Available View Errata Sheet: [Click Here] Download examples: [Click Here] Buy Paperback: $39.99(USD)[Buy] Buy PDF-EBook: $19.99(USD)[Buy] Note: Our PDF books contain no DRM and can be printed, copied to multiple computers owned by you, and once downloaded do not require an internet connection.

Description

Introduction to Neural Networks in Java introduces the Java programmer to the world of Neural Networks and Artificial Intelligence. Neural network architectures such as the feedforward backpropagation, Hopfield, and Kohonen networks are discussed. Additional AI topics, such as Genetic Algorithms and Simulated Annealing, are also introduced.

Practical examples are given for each neural network. Examples include the Traveling Salesman problem, handwriting recognition, fuzzy logic and learning mathematical functions. All Java source code can be downloaded online. In addition to showing the programmer how to construct these neural networks, the book discusses the Java Object Oriented Neural Engine (JOONE). JOONE is a free open source Java neural engine.

Table of Contents

Chapter 1: An Introduction to Neural Networks
The structure of neural networks will be briefly introduced in this chapter. Also discussed is the history of neural networks, since it is important to know where neural networks came from, as well as where they are ultimately headed. Finally, there is a broad overview of both the biological and historic context of neural networks.
Chapter 2: Understanding Neural Networks
A neural network can be trained to recognize specific patterns in data. This chapter will teach you the basic layout of a neural network and end by demonstrating the Hopfield neural network, which is one of the simplest forms of neural network.
Chapter 3: Using Multilayer Neural Networks
You will see how to use the feed-forward multilayer neural network and two ways that you can implement such a neural network. The chapter begins by examining an open source neural network engine called JOONE. JOONE contains a neural network editor that allows you to quickly model and test neural networks.
Chapter 4: How a Machine Learns
Every learning algorithm involves somehow modifying the weight matrices between the neurons. This chapter examines some of the more popular ways of adjusting these weights.
Chapter 5: Understanding Back Propagation
This chapter examines one of the most common neural network architectures-- the feed foreword back propagation neural network.
Chapter 6: Understanding the Kohonen Neural Network
The Kohonen neural network contains no hidden layer. The Kohonen neural network differs from the feedfroward back propagation neural network in several important ways. This chapter examines the Kohonen neural network and how it is implemented.
Chapter 7: Optical Character Recognition
This chapter develops an example program that can be trained to recognize human handwriting. It is not a program that can scan pages of text. Rather this program will read character by character, as the user draws them. This function will be similar to the handwriting recognition used by many PDA's.
Chapter 8: Understanding Genetic Algorithms
A chapter on an AI technology unrelated to neural networks.
Chapter 9: Understanding Simulated Annealing
A second AI technology that can be used to train neural networks.
Chapter 10: Handling Local Minima
One of the most fundamental flaws is the tendency for the backpropagation training algorithm to fall into a "local minima". A local minimum is a false optimal weight matrix that prevents the backpropagation training algorithm from seeing the true solution. This chapter shows how to use certain training techniques to supplement backpropagation and elude local minima.
Chapter 11: Pruning Neural Networks
This chapter examines several algorithms that modify the structure of the neural network. This structural modification will not generally improve the performance of the neural network, but makes it more efficient. If a particular neuron's connection to other neurons does not significantly affect the output of the neural network, the connection will be pruned.
Chapter 12: Fuzzy Logic
Fuzzy logic is a branch of AI not directly related to the neural networks examined so far. Fuzzy logic is often used to process data before it is fed to a neural network, or to process the outputs from the neural network. Fuzzy logic is examined in reference to removing SPAM from emails.
Appendix A: JOONE Reference
Appendix B: Mathematical Backgrounder
Appendix C: Using the Examples on a Windows System
Appendix D: Using the Examples on a UNIX System
]]> Note on .SER files: The book mentions .ser files for use with JOONE, which we no longer include with the download. These are binary images of a JOONE neural network, which seem to break with each new version of JOONE. If you follow the steps in the book, you will easily create the same file, which you can save to disk.
Note on JOONE Versions: JOONE has seen a number of updates since the writing of this book. Namely they have replaced the JOONE editor with JOONEPAD. Heaton Research will be publishing some JOONEPAD articles soon.

Academic and Media Review Copies

If you are considering this book for a class you are teaching, you can [Request a Review Copy]