Neural Networks for Applied Sciences and Engineering: From Fundamentals to Complex Pattern Recognition

ISBN-13: 9780849333750 / Angielski / Twarda / 2006 / 570 str.

Sandhya Samarasinghe

Opis

With the large amounts of data collected by scientists and the advent of faster computers, the demand for novel computing methods for analyzing biological and scientific data is growing exponentially. Addressing this need, Neural Networks for Applied Sciences and Engineering: From Fundamentals to Complex Pattern Recognition provides scientists with a simple but systematic introduction to neural networks. Beginning with an introductory discussion on the role of neural networks in scientific data analysis, this book provides a solid foundation of basic neural network concepts. It contains an overview of neural network architectures for practical data analysis followed by an extensive coverage on linear networks as well as multi-layer networks for nonlinear prediction and clustering with a step-by-step explanation of data processing in these networks and all stages of model development illustrated through practical examples and case studies. This is followed by a detailed treatment of data exploration and preprocessing including dimensionality reduction and input selection, model uncertainty assessment, and sensitivity analysis on inputs, errors and model parameters. Later chapters present an extensive coverage on Self Organizing Maps for data clustering, recurrent networks for time series forecasting and other network types suitable for scientific data analysis. Relevant statistical methods are presented through out the book to demonstrate the complementarities of the two approaches while highlighting the superior nonlinear modeling capabilities of neural networks. In an easy-to-understand format suitable for applied scientists and engineers, this book fills the gap in the market for neural networks tailored to scientific data analysis highlighting all stages of model development. With a multidisciplinary scientific context, it addresses how neural networks perform linear and nonlinear data analysis, including prediction, classifi

Recenzje

"Beginning with the basics, she [Samarasinghe] explains a variety of neural networks' internal workings, and how to apply them to solve real problems."
-SciTech Book News, December 2006

Spis treści

FROM DATA TO MODELS: COMPLEXITY AND CHALLENGES IN UNDERSTANDING BIOLOGICAL, ECOLOGICAL, AND NATURAL SYSTEMS
Introduction
Layout of the Book

FUNDAMENTALS OF NEURAL NETWORKS AND MODELS FOR LINEAR DATA ANALYSIS
Introduction and Overview
Neural Networks and Their Capabilities
Inspirations from Biology
Modeling Information Processing in Neurons
Neuron Models and Learning Strategies
Models for Prediction and Classification
Practical Examples of Linear Neuron Models on Real Data
Comparison with Linear Statistical Methods
Summary
Problems

NEURAL NETWORKS FOR NONLINEAR PATTERN RECOGNITION
Overview and Introduction
Nonlinear Neurons
Practical Example of Modeling with Nonlinear Neurons
Comparison with Nonlinear Regression
One-Input Multilayer Nonlinear Networks
Two-Input Multilayer Perceptron Network
Case Studies on Nonlinear Classification and Prediction with Nonlinear Networks
Multidimensional Data Modeling with Nonlinear Multilayer Perceptron Networks
Summary
Problems

LEARNING OF NONLINEAR PATTERNS BY NEURAL NETWORKS
Introduction and Overview
Supervised Training of Networks for Nonlinear Pattern Recognition
Gradient Descent and Error Minimization
Backpropagation Learning and Illustration with an Example and Case Study
Delta-Bar-Delta Learning and Illustration with an Example and Case Study
Steepest Descent Method Presented with an Example
Comparison of First Order Learning Methods
Second-Order Methods of Error Minimization and Weight Optimization
Comparison of First Order and Second Order Learning Methods Illustrated through an Example
Summary
Problems

IMPLEMENTATION OF NEURAL NETWORK MODELS FOR EXTRACTING RELIABLE PATTERNS FROM DATA
Introduction and Overview
Bias-Variance Tradeoff
Illustration of Early Stopping and Regularization
Improving Generalization of Neural Networks
Network structure Optimization and Illustration with Examples
Reducing Structural Complexity of Networks by Pruning
Demonstration of Pruning with Examples
Robustness of a Network to Perturbation of Weights Illustrated Using an Example
Summary
Problems

DATA EXPLORATION, DIMENSIONALITY REDUCTION, AND FEATURE EXTRACTION
Introduction and Overview
Data Visualization Presented on Example Data
Correlation and Covariance between Variables
Normalization of Data
Example Illustrating Correlation, Covariance and Normalization
Selecting Relevant Inputs
Dimensionality Reduction and Feature Extraction
Example Illustrating Input Selection and Feature Extraction
Outlier Detection
Noise
Case Study: Illustrating Input Selection and Dimensionality Reduction for a
Practical Problem
Summary
Problems

ASSESSMENT OF UNCERTAINTY OF NEURAL NETWORK MODELS USING BAYESIAN STATISTICS
Introduction and Overview
Estimating Weight Uncertainty Using Bayesian Statistics
Case study Illustrating Weight Probability Distribution
Assessing Uncertainty of Neural Network Outputs Using Bayesian Statistics
Case Study Illustrating Uncertainty Assessment of Output Errors
Assessing the Sensitivity of Network Outputs to Inputs
Case Study Illustrating Uncertainty Assessment of Network Sensitivity to Inputs
Summary
Problems

DISCOVERING UNKNOWN CLUSTERS IN DATA WITH SELF-ORGANIZING MAPS
Introduction and Overview
Structure of Unsupervised Networks for Clustering Multidimensional Data
Learning in Unsupervised Networks
Implementation of Competitive Learning Illustrated Through Examples
Self-Organizing Feature Maps
Examples and Case Studies Using Self-Organizing Maps on Multi-Dimensional Data
Map Quality and Features Presented through Examples
Illustration of Forming Clusters on the Map and Cluster Characteristics
Map Validation and an Example
Evolving Self-Organizing Maps
Examples Illustrating Various Evolving Self Organizing Maps
Summary
Problems

NEURAL NETWORKS FOR TIME-SERIES FORECASTING
Introduction and Overview
Linear Forecasting of Time-Series with Statistical and Neural Network Models
Example Case Study
Neural Networks for Nonlinear Time-Series Forecasting
Example Case Study
Hybrid Linear (ARIMA) and Nonlinear Neural Network Models
Example Case Study
Automatic Generation of Network Structure Using Simplest Structure Concept-Illustrated Through Practical Application Case Study
Generalized Neuron Network and Illustration Through Practical Application Case
Study
Dynamically Driven Recurrent Networks
Practical Application Case Studies
Bias and Variance in Time-Series Forecasting Illustrated Through an Example
Long-Term Forecasting and a Case study
Input Selection for Time-Series Forecasting
Case study for Input Selection
Summary
Problems

O autorze

Sandhya Samarasinghe