Icann '93

The field of neural networks is motivated by the commonly held belief that applications in artificial intelligence and robotics will benefit from a good understanding of the neural information processing properties that underlie human intelligence. ICANN '93 presents an overview of research and applications assessing the validity of this belief. Within the field of neural networks, several subfields of research can be distinguished. The first main activity is to extract computational principles from experimental neurobiological findings, such as the information processing capability of neurons, adaptation strategies and neural architectures. The field of cognitive connectionism models higher order brain functions, such as language skills, scene interpretation and reasoning, in terms of networks whose units should be interpreted as large populations of neurons, rather than as single neurons. Through theoretical analysis, the relationship between architecture and function of neural networks is studied, and insight is gained into their information processing capabilities. Neural networks technology leads to practical methods that solve real problems in a wide area of industrial applications. ICANN '93 covers a diversity of fields, from neurobiology and psychological models, to theory of information processing, dynamical systems and generalization, to applications in industry, finance and business. The papers on robotics, pattern recognition, and industrial applications provide an excellent overview of the state of the art in 1993.Gielen, S. is the author of 'Icann '93', published 1993 under ISBN 9783540198390 and ISBN 3540198393.