Ebook: Dealing with Complexity: A Neural Networks Approach

In almost all areas of science and engineering, the use of computers and microcomputers has, in recent years, transformed entire subject areas. What was not even considered possible a decade or two ago is now not only possible but is also part of everyday practice. As a result, a new approach usually needs to be taken (in order) to get the best out of a situation. What is required is now a computer's eye view of the world. However, all is not rosy in this new world. Humans tend to think in two or three dimensions at most, whereas computers can, without complaint, work in n­ dimensions, where n, in practice, gets bigger and bigger each year. As a result of this, more complex problem solutions are being attempted, whether or not the problems themselves are inherently complex. If information is available, it might as well be used, but what can be done with it? Straightforward, traditional computational solutions to this new problem of complexity can, and usually do, produce very unsatisfactory, unreliable and even unworkable results. Recently however, artificial neural networks, which have been found to be very versatile and powerful when dealing with difficulties such as nonlinearities, multivariate systems and high data content, have shown their strengths in general in dealing with complex problems. This volume brings together a collection of top researchers from around the world, in the field of artificial neural networks.

---

This volume brings together a collection of top international researchers in the field of artificial neural networks with the = common theme being an attempt to tackle the problem of complexity. The contributions range from more theoretical analyses of the neural network approach to a number of application-oriented articles which indicate the extent of the problem from a more practical viewpoint. The use of neural networks is a relatively new, but increasingly popular, approach to the problem of complexity. Dealing with Complexity is an extremely multi-disciplinary = examination of the above issues: although primarily intended for industrial/academic researchers, and postgraduate students working within computing science, it will also be of interest to anyone=20 working on relevant research projects or applications within the following fields: physics, mathematics and engineering.

---

This volume brings together a collection of top international researchers in the field of artificial neural networks with the = common theme being an attempt to tackle the problem of complexity. The contributions range from more theoretical analyses of the neural network approach to a number of application-oriented articles which indicate the extent of the problem from a more practical viewpoint. The use of neural networks is a relatively new, but increasingly popular, approach to the problem of complexity. Dealing with Complexity is an extremely multi-disciplinary = examination of the above issues: although primarily intended for industrial/academic researchers, and postgraduate students working within computing science, it will also be of interest to anyone=20 working on relevant research projects or applications within the following fields: physics, mathematics and engineering.
Content:
Front Matter....Pages I-XV
Recurrent Neural Networks: Some Systems-Theoretic Aspects....Pages 1-12
The Use of State Space Control Theory for Analysing Feedforward Neural Networks....Pages 13-28
Statistical Decision Making and Neural Networks....Pages 29-46
A Tutorial on the EM Algorithm and Its Applications to Neural Network Learning....Pages 47-61
On the Effectiveness of Memory-Based Methods in Machine Learning....Pages 62-75
A Study of Non Mean Square Error Criteria for the Training of Neural Networks....Pages 76-92
A Priori Information in Network Design....Pages 93-109
Neurofuzzy Systems Modelling: A Transparent Approach....Pages 110-125
Feature Selection and Classification by a Modified Model with Latent Structure....Pages 126-140
Geometric Algebra Based Neural Networks....Pages 141-157
Discrete Event Complex Systems: Scheduling with Neural Networks....Pages 158-176
Incremental Approximation by Neural Networks....Pages 177-188
Approximation of Smooth Functions by Neural Networks....Pages 189-204
Rates of Approximation in a Feedforward Network Depend on the Type of Computational Unit....Pages 205-219
Recent Results and Mathematical Methods for Functional Approximation by Neural Networks....Pages 220-237
Differential Neurocontrol of Multidimensional Systems....Pages 238-251
The Psychological Limits of Neural Computation....Pages 252-263
A Brain-Like Design to Learn Optimal Decision Strategies in Complex Environments....Pages 264-303
Back Matter....Pages 304-308

---

This volume brings together a collection of top international researchers in the field of artificial neural networks with the = common theme being an attempt to tackle the problem of complexity. The contributions range from more theoretical analyses of the neural network approach to a number of application-oriented articles which indicate the extent of the problem from a more practical viewpoint. The use of neural networks is a relatively new, but increasingly popular, approach to the problem of complexity. Dealing with Complexity is an extremely multi-disciplinary = examination of the above issues: although primarily intended for industrial/academic researchers, and postgraduate students working within computing science, it will also be of interest to anyone=20 working on relevant research projects or applications within the following fields: physics, mathematics and engineering.
Content:
Front Matter....Pages I-XV
Recurrent Neural Networks: Some Systems-Theoretic Aspects....Pages 1-12
The Use of State Space Control Theory for Analysing Feedforward Neural Networks....Pages 13-28
Statistical Decision Making and Neural Networks....Pages 29-46
A Tutorial on the EM Algorithm and Its Applications to Neural Network Learning....Pages 47-61
On the Effectiveness of Memory-Based Methods in Machine Learning....Pages 62-75
A Study of Non Mean Square Error Criteria for the Training of Neural Networks....Pages 76-92
A Priori Information in Network Design....Pages 93-109
Neurofuzzy Systems Modelling: A Transparent Approach....Pages 110-125
Feature Selection and Classification by a Modified Model with Latent Structure....Pages 126-140
Geometric Algebra Based Neural Networks....Pages 141-157
Discrete Event Complex Systems: Scheduling with Neural Networks....Pages 158-176
Incremental Approximation by Neural Networks....Pages 177-188
Approximation of Smooth Functions by Neural Networks....Pages 189-204
Rates of Approximation in a Feedforward Network Depend on the Type of Computational Unit....Pages 205-219
Recent Results and Mathematical Methods for Functional Approximation by Neural Networks....Pages 220-237
Differential Neurocontrol of Multidimensional Systems....Pages 238-251
The Psychological Limits of Neural Computation....Pages 252-263
A Brain-Like Design to Learn Optimal Decision Strategies in Complex Environments....Pages 264-303
Back Matter....Pages 304-308
....