Measures of complexity. Proceedings of the conference, held in Rome, September 30 - October 2, 1987 (Q1210905)

[The articles of this volume will not be indexed individually.] According to the preface, the question asked to the speakers in the conference was to identify a manner of quantitatively characterizing complexity within each of their own disciplines and - if possible - to compare it with the corresponding ones of other disciplines. There are 12 contributions from various disciplines. - P. Grassberger defines that the complexity of an object is the difficulty of the most important task associated with this object. He considers sequences of a finite number of symbols of some alphabet and defines three categories of the complexity: regular language complexity, set complexity and proper forecasting complexity. G. Parisi considers various possibilities to define the complexity and points out examples from the equilibrium statistical mechanics. Ph. de Forcrand, F. Koukiou and D. Petritis distinguish in their report about random walks and random surfaces the mathematical, statistical and algorithmic complexity. D. P. Bovet and P. L. Crescenzi define the measure of complexity of Turing machine computations. They distinguish static and dynamic complexity. The measure of the dynamic complexity can be either the number of steps or the number of cells scanned during the computation. The complexity in large technological systems and the complexity of biological and ecological systems are also discussed. Some authors regard the complexity as the difficulty in understanding a system. Some reports describe complex systems rather than define the measure of the complexity (e.g. several reports about learning processes).