CALL FOR CHAPTERS

Proposals Submission Deadline: 10/31/2008
Full Chapters Due: 02/28/2009

Artificial Higher Order Neural Networks for Computer Science and Engineering:
Trends for Emerging Applications

A book edited by Prof. Ming Zhang (PhD), Christopher Newport University, VA, USA

Introduction
Artificial Neural Networks (ANNs) are known to excel in pattern recognition, pattern matching and mathematical function approximation. However they suffer from several well known limitations– they can often become stuck in local, rather than global minima, as well as taking an unacceptable amount of time to converge in practice. Of particular concern, especially from the perspective of economics and financial time series predictions, is their inability to handle non-smooth, discontinuous training data, and complex mappings (associations). Another limitation of ANN is a ‘black box’ nature – meaning that explanations (reasons) for their decisions are not immediately obvious, unlike techniques such as Decision Trees. This then is the motivation for developing artificial Higher Order Neural Networks (HONNs), since HONNs are ‘open-box’ models and each neuron and weight are mapped to function variable and coefficient.

In recent years, researchers use HONNs for pattern recognition, nonlinear simulation, classification, and prediction in the computer science and computer engineering areas. The results show that HONNs are always faster, more accurate, and easier to explain. This is the second motivation for using HONNs in computer science and computer engineering areas, since HONNs can automatically select the initial coefficients, and can even automatically select the model for applications in computer science and computer engineering.

Giles & Maxwell (1987) published the first paper on HONN. Bengtsson (1990) wrote the first book in the higher order (or higher-order, consistency) neural network area. Higher order correlations in the training data require more complex neuron activation functions (Barron, Gilstrap & Shrier, 1987; Giles & Maxwell, 1987; Psaltis, Park & Hong, 1988). Neurons which include terms up to and including degree-k are referred to as kth-order neurons (Lisboa & Perantonis ,1991). Currently the output of a kth-order single-layer HONN neuron will be a non-linear function comprising polynomials of up to kth-order. Moreover, since no hidden layers are involved, both Hebbian and Perceptron learning rules can be employed (Shin & Ghosh, 1991). The Neuron-Adaptive HONN (and NAHONN group) leads to faster convergence, much reduced network size and more accurate curve fitting, compared with P(T)HONNs (Zhang, Xu & Fulcher, 2002).A more comprehensive coverage, including derivations of weight update equations, is presented in Zhang & Fulcher (2004). Now as with the earlier HONN groups, it is possible to provide a similar general result to that found previously by Hornik (1991) for ANNs – namely that NAHONN groups are capable of approximating any kind of piecewise continuous function, to any degree of accuracy (a proof is provided in Zhang, Xu & Fulcher, 2002). Moreover, these models are capable of automatically selecting not only the optimum model for a particular time series, but also the appropriate model order and coefficients.
The Objectives and Mission of the Book
(1) Objectives

• This is the first book which introduces and explains HONNs to people working in the fields of computer science and computer engineering, and how to use HONNS in these areas. HONN is an open box neural networks tool compared to traditional artificial neural networks.
• This is the first book which includes the most popular HONNs software packages and detailed information for researchers to successfully use these HONNs software packages.
• This book explains why HONNs can approximate any nonlinear data to any degree of accuracy, and allows researchers to understand why HONNs are much easier to use, and HONNs can have better nonlinear data simulation accuracy than SAS nonlinear (NLIN) procedures. This book introduces the HONN group models and adaptive HONNs, and allows the people working in the computer science and computer engineering areas to understand HONN group models and adaptive HONN models, which can simulate not only nonlinear data, but also discontinuous and unsmooth nonlinear data.

(2) Mission

Inform researchers and practitioners working in the computer science and computer engineering areas that HONNs are much easier to use and can have better simulation results than SAS Nonlinear models, and to illustrate how to successfully use HONNs software packages and hardware designs for nonlinear data simulation and prediction. HONNs will challenge traditional artificial neural network products and change the research methodology that people are currently using in computer science and computer engineering areas for the pattern recognition, nonlinear simulation, classification, and prediction.

The Target Audience
Researchers and practitioners who are using artificial neural networks and who are doing nonlinear model research, in particular, professors, graduate students, and senior undergraduate students in the computer science and computer engineering departments, as well as the professionals and researchers in these areas.
Recommended topics include, but are not limited to, the following areas:
1. Artificial Higher Order Neural Network Models
2. Artificial Higher Order Neural Network Software
3. Artificial Higher Order Neural Network for Pattern Recognition
4. Artificial Higher Order Neural Networks for Computer Science and Engineering
5. Artificial Higher Order Neural Networks for Other Areas

More detailed list of topics is available upon request.
Submission Process
Researchers and practitioners are invited to submit on or before October 31, 2008, a 2-5 page (600-1500 word) manuscript proposal clearly explaining the mission and concepts of the proposed work. Proposals should be sent to mzhang@pcs.cnu.edu. Submitters will be notified by November 30, 2008, about the status of their proposals. Authors of accepted proposals will be sent organizational guidelines. Completed chapters are expected to be submitted by February 28, 2009. Submitted essays will be double-blind reviewed. The book is scheduled to be published in 2009 by IGI Global Inc., www.igi-global.com, publisher of the Information Science Reference (formerly Idea Group Reference) and Medical Information Science Reference imprints.
Inquiries and submissions can be forwarded electronically (Word document) or by mail to:
Prof. Ming Zhang (PhD)
Department of Physics, Computer Science and Engineering
Christopher Newport University
1 University Place, Newport News, VA 23606, USA
Phone: 1-757 594 7563; Fax: 1-757 594 7919
E-mail: mzhang@pcs.cnu.edu