This book constitutes the refereed proceedings of the sixth International Conference on Artificial Neural Networks - ICANN 96, held in Bochum, Germany in July 1996.
The 145 papers included were carefully selected from numerous submissions on the basis of at least three reviews; also included are abstracts of the six invited plenary talks. All in all, the set of papers presented reflects the state of the art in the field of ANNs. Among the topics and areas covered are a broad spectrum of theoretical aspects, applications in various fields, sensory processing, cognitive science and AI, implementations, and neurobiology.

The topic of consciousness is truly multidisciplinary, attracting researchers and theorists from diverse backgrounds. It is now widely accepted that previously disparate areas all have contributions to make to the understanding of the nature of consciousness. Thus, we now have computational scientists, neuroscientists, and philosophers all engaged in the same effort. This book illustrates these three approaches, with chapters provided by some of the most important and provocative figures in the field. The first section is concerned with philosophical approaches to consciousness. One of the fundamental issues here is that of subjective feeling or qualia. The second section focuses on approaches from cognitive neuroscience. Patients with different types of neurological problems, and new imaging techniques, provide rich sources of data for studying how consciousness relates to brain function. The third section includes computational approaches looking at the quantitative relationship between brain processes and conscious experience. Cognition, Computation, and Consciousness represents a uniquely integrated and current account of this most fascinating and intractable subject.

This volume presents a collection of revised refereed papers selected from the contributions presented at the European Conference on Artificial Evolution, AE '95, held in Brest, France, in September 1995; also included are a few papers from the predecessor conference, AE '94.
Besides two invited surveys on evolution strategies and evolutionary programming, 24 full papers are presented. They are grouped into sections on evolutionary computation theory, genetic algorithm techniques, coevolution, neural networks, image processing, and applications to various optimization and other problems.

This book presents 14 rigorously reviewed revised papers selected from more than 50 submissions for the 1994 IEEE/ Nagoya-University World Wisepersons Workshop, WWW'94, held in August 1994 in Nagoya, Japan.The combination of approaches based on fuzzy logic, neural networks and genetic algorithms are expected to open a new paradigm of machine learning for the realization of human-like information processing systems. The first six papers in this volume are devoted to the combination of fuzzy logic and neural networks; four papers are on how to combine fuzzy logic and genetic algorithms. Four papers investigate challenging applications of fuzzy systems and of fuzzy-genetic algorithms.

Neural Networks presents concepts of neural-network models and techniques of parallel distributed processing in a three-step approach: - A brief overview of the neural structure of the brain and the history of neural-network modeling introduces to associative memory, preceptrons, feature-sensitive networks, learning strategies, and practical applications. - The second part covers subjects like statistical physics of spin glasses, the mean-field theory of the Hopfield model, and the "space of interactions" approach to the storage capacity of neural networks. - The final part discusses nine programs with practical demonstrations of neural-network models. The software and source code in C are on a 3 1/2" MS-DOS diskette can be run with Microsoft, Borland, Turbo-C, or compatible compilers.

A learning system can be defined as a system which can adapt its behaviour to become more effective at a particular task or set of tasks. It consists of an architecture with a set of variable parameters and an algorithm. Learning systems are useful in many fields, one of the major areas being in control and system identification. This work covers major aspects of learning systems: system architecture, choice of performance index and methods measuring error. Major learning algorithms are explained, including proofs of convergence. Artificial neural networks, which are an important class of learning systems and have been subject to rapidly increasing popularity, are discussed. Where appropriate, examples have been given to demonstrate the practical use of techniques developed in the text. System identification and control using multi-layer networks and CMAC (Cerebellar Model Articulation Controller) are also presented.

Neural networks is a field of research which has enjoyed rapid expansion in both the academic and industrial research communities. This volume contains papers presented at the Third Annual SNN Symposium on Neural Networks to be held in Nijmegen, The Netherlands, 14 - 15 September 1995. The papers are divided into two sections: the first gives an overview of new developments in neurobiology, the cognitive sciences, robotics, vision and data modelling. The second presents working neural network solutions to real industrial problems, including process control, finance and marketing. The resulting volume gives a comprehensive view of the state of the art in 1995 and will provide essential reading for postgraduate students and academic/industrial researchers.

This book presents carefully revised versions of tutorial lectures given during a School on Artificial Neural Networks for the industrial world held at the University of Limburg in Maastricht, Belgium.
The major ANN architectures are discussed to show their powerful possibilities for empirical data analysis, particularly in situations where other methods seem to fail. Theoretical insight is offered by examining the underlying mathematical principles in a detailed, yet clear and illuminating way. Practical experience is provided by discussing several real-world applications in such areas as control, optimization, pattern recognition, software engineering, robotics, operations research, and CAM.

This volume contains the thoroughly refereed and revised papers accepted for presentation at the IJCAI '91 Workshops on Fuzzy Logic and Fuzzy Control, held during the International Joint Conference on AI at Sydney, Australia in August 1991. The 14 technical contributions are devoted to several theoretical and applicational aspects of fuzzy logic and fuzzy control; they are presented in sections on theoretical aspects of fuzzy reasoning and fuzzy control, fuzzy neural networks, fuzzy control applications, fuzzy logic planning, and fuzzy circuits. In addition, there is a substantial introduction by the volume editors on the latest developments in the field that brings the papers presented into line.

This three-part monograph addresses topics in the areas of control systems, signal processing and neural networks. Procedures and results are determined which constitute the first successful synthesis procedure for associative memories by means of artificial neural networks with arbitrarily pre-specified full or partial interconnecting structure and with or without symmetry constraints for the connection matrix.

Artificial neural networks and genetic algorithms both are areas of research which have their origins in mathematical models constructed in order to gain understanding of important natural processes. By focussing on the process models rather than the processes themselves, significant new computational techniques have evolved which have found application in a large number of diverse fields. This diversity is reflected in the topics which are subjects of the contributions to this volume. There are contributions reporting successful applications of the technology to the solution of industrial/commercial problems. This may well reflect the maturity of the technology, notably in the sense that 'real' users of modelling/prediction techniques are prepared to accept neural networks as a valid paradigm. Theoretical issues also receive attention, notably in connection with the radial basis function neural network. Contributions in the field of genetic algorithms reflect the wide range of current applications, including, for example, portfolio selection, filter design, frequency assignment, tuning of nonlinear PID controllers. These techniques are also used extensively for combinatorial optimisation problems.

This volume contains the proceedings of the 15th International Conference on Application and Theory of Petri Nets, held at Zaragoza, Spain in June 1994. The annual Petri net conferences are usually visited by some 150 - 200 Petri net experts coming from academia and industry all over the world.
The 28 full papers were selected out of in total 85 submissions from 18 countries. Together with the 3 invited contributions, these papers give a state-of-the-art report on all relevant aspects of Petri nets.

This volume contains the papers that were presented at the Third Workshop onAlgorithmic Learning Theory, held in Tokyo in October 1992. In addition to 3invited papers, the volume contains 19 papers accepted for presentation, selected from 29 submitted extended abstracts. The ALT workshops have been held annually since 1990 and are organized and sponsored by the Japanese Society for Artificial Intelligence. The main objective of these workshops is to provide an open forum for discussions and exchanges of ideasbetween researchers from various backgrounds in this emerging, interdisciplinary field of learning theory. The volume is organized into parts on learning via query, neural networks, inductive inference, analogical reasoning, and approximate learning.

From its early beginnings in the fifties and sixties the field of neural networks has been steadily growing. The first wave was driven by a handful of pioneers who first discovered analogies between machines and biological systems in communication, control and computing. Technological constraints held back research considerably, but gradually computers have become less expensive and more accessible and software tools inceasingly more powerful. Mathematical techniques, developed by computer-aware people, have steadily accumulated and the second wave has begun. Researchers from such diverse areas as psychology, mathematics, physics, neuroscience and engineering now work together in the neural networking field.

The Central Nervous System can be considered as an aggregate of neurons specialized in both the transmission and transformation of information. Information can be used for many purposes, but probably the most important one is to generate a representation of the "external" world that allows the organism to react properly to changes in its external environment. These functions range from such basic ones as detection of changes that may lead to tissue damage and eventual destruction of the organism and the implementation of avoidance reactions, to more elaborate representations of the external world implying recognition of shapes, sounds and textures as the basis of planned action or even reflection. Some of these functions confer a clear survival advantage to the organism (prey or mate recognition, escape reactions, etc. ). Others can be considered as an essential part of cognitive processes that contribute, to varying degrees, to the development of individuality and self-consciousness. How can we hope to understand the complexity inherent in this range of functionalities? One of the distinguishing features of the last two decades has been the availability of computational power that has impacted many areas of science. In neurophysiology, computation is used for experiment control, data analysis and for the construction of models that simulate particular systems. Analysis of the behavior of neuronal networks has transcended the limits of neuroscience and is now a discipline in itself, with potential applications both in the neural sciences and in computing sciences.

This book is the product of a 15-month intensive investigation of the European artificial network scene, together with a view of the broader framework of the subject in a world context. It could not have been completed in such a remarkably short time, and so effectively, without the dedicated efforts of Louise Turner, the DEANNA secretary, and Geoff Chappell, the DEANNA researcher, at the Centre for Neural Networks, King's College, London. I would like to take this opportunity to thank them for their heroic efforts. I would also like to thank my colleagues in the Centre and in the Mathematics Department, especially Mark Plumbley, Michael Reiss and Trevor Clarkson for all their help and encouragement, Denise Gorse of University College London, for allowing use of her lecture notes as a basis for the tutorial and the DEANNA partners for the part they played. Finally I would like to acknowledge the European Community support, and especially Mike Coyle for his trenchant comments during the carrying out of the work. March 1993 J. G. Taylor CONTENTS PART I: SETTING THE SCENE Chapter 1: DEANNA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1 . 1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 . 2 The Geographical Dimension. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1 1. 3 The Industrial Dimension. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1 . 4 The Plan for Neural Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Chapter 2: Neural Net Demonstrators. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2. 1 The Status of Neural Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2. 2 Reasons for the Employment of Neural Networks . . . . . . . . . . . . . . . . . . . 9 2. 3 Neural Network Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2. 4 Areas of Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2. 5 Typical Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

The problem of designing a cost-efficient network that survives the failure of one or more nodes or edges of the network is critical to modern telecommunications engineering. The method developed in this book is designed to solve such problems to optimality. In particular, a cutting plane approach is described, based on polyhedral combinatorics, that is ableto solve real-world problems of this type in short computation time. These results are of interest for practitioners in the area of communication network design. The book is addressed especially to the combinatorial optimization community, but also to those who want to learn polyhedral methods. In addition, interesting new research problemsare formulated.

This textbook gives details of recent developments in the field of image processing, machine vision and analysis. Based on the original book published in Czech, this English edition has been expanded to include 3D vision, neural networks and invariants.

This book contains the proceedings of the International Confer ence on Artificial Neural Networks which was held between September 13 and 16 in Amsterdam. It is the third in a series which started two years ago in Helsinki and which last year took place in Brighton. Thanks to the European Neural Network Society, ICANN has emerged as the leading conference on neural networks in Europe. Neural networks is a field of research which has enjoyed a rapid expansion and great popularity in both the academic and industrial research communities. The field is motivated by the commonly held belief that applications in the fields of artificial intelligence and robotics will benefit from a good understanding of the neural information processing properties that underlie human intelligence. Essential aspects of neural information processing are highly parallel execution of com putation, integration of memory and process, and robustness against fluctuations. It is believed that intelligent skills, such as perception, motion and cognition, can be easier realized in neuro-computers than in a conventional computing paradigm. This requires active research in neurobiology to extract com putational principles from experimental neurobiological find ings, in physics and mathematics to study the relation between architecture and function in neural networks, and in cognitive science to study higher brain functions, such as language and reasoning. Neural networks technology has already lead to practical methods that solve real problems in a wide area of industrial applications. The clusters on robotics and applications contain sessions on various sub-topics in these fields."

Neural network technology encompasses a class of methods which attempt to mimic the basic structures used in the brain for information processing. Thetechnology is aimed at problems such as pattern recognition which are difficult for traditional computational methods. Neural networks have potential applications in many industrial areas such as advanced robotics, operations research, and process engineering. This book is concerned with the application of neural network technology to real industrial problems. It summarizes a three-year collaborative international project called ANNIE (Applications of Neural Networks for Industry in Europe) which was jointly funded by industry and the European Commission within the ESPRIT programme. As a record of a working project, the book gives an insight into the real problems faced in taking a new technology from the workbench into a live industrial application, and shows just how it can be achieved. It stresses the comparison between neural networks and conventional approaches. Even the non-specialist reader will benefit from understanding the limitations as well as the advantages of the new technology.

Neural Network Dynamics is the latest volume in the "Perspectives in Neural Computing" series. It contains papers presented at the 1991 Workshop on Complex Dynamics in Neural Networks, held at IIASS in Vietri, Italy. The workshop encompassed a wide range of topics in which neural networks play a fundamental role, and aimed to bridge the gap between neural computation and computational neuroscience. The papers - which have been updated where necessary to include new results - are divided into four sections, covering the foundations of neural network dynamics, oscillatory neural networks, as well as scientific and biological applications of neural networks. Among the topics discussed are: A general analysis of neural network activity; Descriptions of various network architectures and nodes; Correlated neuronal firing; A theoretical framework for analyzing the behaviour of real and simulated neuronal networks; The structural properties of proteins; Nuclear phenomenology; Resonance searches in high energy physics; The investigation of information storage; Visual cortical architecture; Visual processing. Neural Network Dynamics is the first volume to cover neural networks and computational neuroscience in such detail. Although it is primarily aimed at researchers and postgraduate students in the above disciplines, it will also be of interest to researchers in electrical engineering, medicine, psychology and philosophy.

The purpose of this book is to develop neural nets as a strong theory for both brains and machines. The theory is developed in close correlation with the biology of the neuron and the properties of human reasoning. This approach implies the following: - Updating the biology of the artificialneuron. The neurosciences have experienced a tremendous development in the last 50 years. One of the main purposes of the present work is toincorporate this knowledge into a strong model of the artificial neuron. Particular attention is devoted to formalizing the complex chemical processes at the synaptic level. This formal language supports both symbolicreasoning and uncertainty processing. - Investigating the properties of expert reasoning. This kind of reasoning is approximate, partial and non-monotonic, and therefore requires special mathematical tools for its formalization, such as fuzzy set theory and fuzzy logic. Three different intelligent systems developed with this technology are presented and discussed.

Neural Network Applications contains the 12 papers presented at the second British Neural Network Society Meeting (NCM '91) held at King's College London on 1st October 1991. The meeting was sponsored by the Centre for Neural Networks, King's College, and the British Neural Network Society, and was also part of the DEANNA ESPRIT programme. The papers reflect the wide spectrum of neural network applications that are currently being attempted in industry and medicine. They cover medical diagnosis, robotics, plant control, machine learning, and visual inspection, as well as more general discussions on net learning and knowledge representation. The breadth and depth of coverage is a sign of the health of the subject, as well as indicating the importance of neural network developments in industry and the manner in which the applications are progressing. Among the actual topics covered are: Learning algorithms - theory and practice; A review of medical diagnostic applications of neural networks; Simulated ultrasound tomographic imaging of defects; Linear quadtrees for neural network based position invariant pattern recognition; The pRTAM as a hardware-realisable neuron; The cognitive modalities ("CM") system of knowledge representation - the DNA of neural networks? This volume provides valuable reading for all those attempting to apply neural networks, as well as those entering the field, including researchers and postgraduate students in computational neuroscience, neurobiology, electrical engineering, computer science, mathematics, and medicine.

Artificial neural networks and genetic algorithms both are areas of research which have their origins in mathematical models constructed in order to gain understanding of important natural processes. By focussing on the process models rather than the processes themselves, significant new computational techniques have evolved which have found application in a large number of diverse fields. This diversity is reflected in the topics which are the subjects of contributions to this volume. There are contributions reporting theoretical developments in the design of neural networks, and in the management of their learning. In a number of contributions, applications to speech recognition tasks, control of industrial processes as well as to credit scoring, and so on, are reflected. Regarding genetic algorithms, several methodological papers consider how genetic algorithms can be improved using an experimental approach, as well as by hybridizing with other useful techniques such as tabu search. The closely related area of classifier systems also receives a significant amount of coverage, aiming at better ways for their implementation. Further, while there are many contributions which explore ways in which genetic algorithms can be applied to real problems, nearly all involve some understanding of the context in order to apply the genetic algorithm paradigm more successfully. That this can indeed be done is evidenced by the range of applications covered in this volume.