Maximum entropy, information without probability and complex fractals. Classical and quantum approach (Q1594603): Difference between revisions

The present book offers many original ideas and developments extending the view on entropy and its application to random processes. In the first half of the book (Chapters 1-5) the author develops the concept of entropy of non-random functions and non-probabilistic matrices, which he also calls path entropies. This latter name is less misleading than the former since measures of uncertainty are used in the construction which make no sense without randomness. For \(C^1({\mathbb{R}}^n,{\mathbb{R}}^n)\)-functions this measure is determined by the modulus of the determinant of the Jacobian. Then the path entropies are derived by application of the maximum entropy principle as formulated by \textit{E. T. Janes} [Phys. Rev. 106, 620-630 (1957); Phys Rev. 108, 171-190 (1957)]. Path entropies are derived for the Shannon, Renyi, and structural entropies. Also for functions depending on a random parameter and stochastic processes, path entropies are derived in this manner. A new set of axioms allows the author to derive the v. Neumann entropy and the Shannon entropy on the same axiomatic background. The quantum entropy of non-probabilistic matrices has no analogous meaning to that of non-random functions as the reader may expect at first sight: The measure of uncertainty used restricts it mainly to density matrices and seems to have no meaning for observables. Nevertheless, general applications concerning complexity, relative entropy, quantum divergencies, and, eventually, differential operators are sketched. The second half of the book concerns complex valued fractional Brownian motions of order \(n\) and applications. In Chapter 6 such processes are defined and modeled in different ways. The relation of Brownian motion to white noise is considered and several properties are described. In Chapter 7 several features of the theory are worked out, especially the adaption of an Ito-calculus for such processes is given. Chapter 8 is devoted to applications with respect to thermodynamics, statistics, and open systems. It is shown how the concept of path entropies introduced in the first half of the book can be applied with advantage to describe the solutions. In Chapter 9 this is especially worked out for stochastic processes of fractional order. An outline of further applications to the stochastic stability of dynamical systems, time series of fractional order, and master equations of fractional order is given in Chapter 10. This book presents a rich spectrum of original ideas and developments useful for theoretical derivations and applications.