Gossiping Nets

The question of how the human mind works and what the basis of its admirable capabilities is probably goes back to the first human who thought about how the very act of thinking comes about. In the meantime, psychologists have acquired plenty of material on human behavior, learning, and perception, while physiologists have uncovered an immense collection of properties of nerve cells and the structures they form in the brain. Starting with a paper by McCulloch and Pitts [1] these biological findings have been abstracted to a series of simplified models, which can be defined in mathematical terms and studied in their own right. According to this model, the brain is “nothing but” a set of simple elements interconnected to a huge graph with some laws governing the time course of the distribution of activity across the elements. Some of the elements are connected to sensors like eyes or ears and are considered input, others are attached to muscles and are considered output. For simplicity, the elements are called neurons, the graphs they form “neural networks” or “neural nets”. Since 1946, this basic idea has been followed, refined, tested and put to use. It turned out that the “nothing but” is a huge euphemism, both because the model is much too simple to cover the biological properties of a brain and even small neural networks can be extremely complicated systems. As for the question of how the mind works, many more open questions have been raised than definitive answers given. Formalized systems have been studied in considerable detail and some of them have become standard tools in statistics, control theory, and computer science. Researchers working on this idea found themselves in the situation that no academic discipline was really responsible for their research and some even rejected it quite fiercely. Consequently, the now established research discipline of Neurocomputing continues to have strong influences from electrical engineering, mathematics, computer science, psychology, neurobiology, and physiology. Clearly the development of such a field cannot proceed without tension.