摘要: Since the time of McCulloch and Pitts’ Theory (1943) there have been many attempts to model flow activity in neural networks. It is possible simulate networks (of rather small size) on a computer, relying quite reasonable — more or less simplified assumptions dynamic behavior single neurons. One problem arbitrariness design network (i.e. connectivity matrix). Here investigations studied random (e.g. Anninos et al. 1970, Griffith 1971, Amari 1974, Dammasch Wagner 1984) that itself changes subject certain rules (for an overview see Palm 1982).