Portal de Periódicos da CAPES

Intensive entropic non-triviality measure

2003; Elsevier BV; Volume: 334; Issue: 1-2 Linguagem: Inglês

10.1016/j.physa.2003.11.005

1873-2119

Pedro W. Lamberti, María T. Martín, A. Plastino, Osvaldo A. Rosso,

Advanced Thermodynamics and Statistical Mechanics

We discuss a way of characterizing probability distributions, complementing that provided by the celebrated notion of information measure, with reference to a measure of complexity that we call a “nontriviality measure”. Our starting point is the “LMC” measure of complexity advanced by López-Ruiz et al. (Phys. Lett. A 209 (1995) 321) and its analysis by Anteneodo and Plastino (Phys. Lett. A 223 (1997) 348). An improvement of some of their troublesome characteristics is thereby achieved. Basically, we replace the Euclidean distance to equilibrium by the Jensen–Shannon divergence. The resulting measure turns out to be (i) an intensive quantity and (ii) allows one to distinguish between different degrees of periodicity. We apply the “cured” measure to the logistic map so as to clearly exhibit its advantages.