Search PubMed⌕ Search

Biomedical subjects

Constantino Tsallis

Publications and source records attributed to Constantino Tsallis.

12 recordsLinked to original sources

Chaos edges of z-logistic maps: connection between the relaxation and sensitivity entropic indices.

Chaos thresholds of the z-logistic maps x(t+1)=1-a|xt|(z) (z>1; t=0,1,2,...) are numerically analyzed at accumulation points of cycles 2, 3, and 5 (three different cycles 5). We verify that the nonextensive q-generalization of a Pesin-like identity is preserved through averaging over the entire phase space. More precisely, we computationally verify lim(t-->infinity) [formula-see text], where the entropy S(q) [formula-see text], the sensitivity to the initial conditions xi(triple bond)lim(Deltax(0)-->0)Deltax(t)/Deltax(0), and ln(q)x(triple bond)(x(1-q-1/(1-q)(ln(1)x=ln x). The entropic index [formula-see text], and the coefficient [formula-see text] depend on both z and the cycle. We also study the relaxation that occurs if we start with an ensemble of initial conditions homogeneously occupying the entire phase space. The associated Lebesgue measure asymptotically decreases as 1/t(1/q(rel)-1)(q(rel>1). These results (i) illustrate the connection (conjectured by one of us) between sensitivity and relaxation entropic indices, namely, [formula-see text], where the positive numbers Alpha(n), alpha(n) depend on the cycle; (ii) exhibit an unexpected scaling, namely, [formula-see text].

Journal Article↗

Generative model for feedback networks.

We propose a model for network formation and study some of its statistical properties. The motivation for the model comes from the growth of several kinds of real networks (i.e., kinship and trading networks, networks of corporate alliances, networks of autocatalytic chemical reactions). These networks grow either by establishing closer connections by adding links in the existing network or by adding new nodes. A node in these networks lacks the information of the entire network. In order to establish a closer connection to other nodes it starts a search in the neighboring part of the network and waits for a possible feedback from a distant node that received the "searching signal." Our model imitates this behavior by growing the network via the addition of a link that creates a cycle in the network or via the addition of a new node with a link to the network. The forming of a cycle creates feedback between the two ending nodes. After choosing a starting node, a search is made for another node at a suitable distance; if such a node is found, a link is established between this and the starting node, otherwise (such a node cannot be found) a new node is added and is linked to the starting node. We simulate this algorithm and find that we cannot reject the hypothesis that the empirical degree distribution is a q-exponential function, which has been used to model long-range processes in nonequilibrium statistical mechanics.

Journal Article↗

Asymptotically scale-invariant occupancy of phase space makes the entropy Sq extensive.

Phase space can be constructed for N equal and distinguishable subsystems that could be probabilistically either weakly correlated or strongly correlated. If they are locally correlated, we expect the Boltzmann-Gibbs entropy S(BG) identical with -k Sigma(i) p(i) ln p(i) to be extensive, i.e., S(BG)(N) proportional, variant N for N --> infinity. In particular, if they are independent, S(BG) is strictly additive, i.e., S(BG)(N) = NS(BG)(1), for allN. However, if the subsystems are globally correlated, we expect, for a vast class of systems, the entropy S(q) identical with k[1 - Sigma(i) p(q)(i)]/(q - 1) (with S(1) = S(BG)) for some special value of q not equal 1 to be the one which is extensive [i.e., S(q)(N) proportional, variant N for N --> infinity]. Another concept which is relevant is strict or asymptotic scale-freedom (or scale-invariance), defined as the situation for which all marginal probabilities of the N-system coincide or asymptotically approach (for N --> infinity) the joint probabilities of the (N - 1)-system. If each subsystem is a binary one, scale-freedom is guaranteed by what we hereafter refer to as the Leibnitz rule, i.e., the sum of two successive joint probabilities of the N-system coincides or asymptotically approaches the corresponding joint probability of the (N - 1)-system. The kinds of interplay of these various concepts are illustrated in several examples. One of them justifies the title of this paper. We conjecture that these mechanisms are deeply related to the very frequent emergence, in natural and artificial complex systems, of scale-free structures and to their connections with nonextensive statistical mechanics. Summarizing, we have shown that, for asymptotically scale-invariant systems, it is S(q) with q not equal 1, and not S(BG), the entropy which matches standard, clausius-like, prescriptions of classical thermodynamics.

Journal Article↗

Generalized entropy arising from a distribution of q indices.

It is by now well known that the Boltzmann-Gibbs (BG) entropy can be usefully generalized using the non-extensive entropies, which have been applied to a wide range of phenomena. However, it seems that even more general entropies could be useful in order to describe other complex physical systems, a task which has already been undertaken in the literature. Following this approach, we introduce here a quite general entropy based on a distribution of q indices thus generalizing S(q). We establish some general mathematical properties for the new entropic functional and explore some examples. We also exhibit a procedure for finding, given any entropic functional, the q-indices distribution that produces it. Finally, on the road to establishing a quite general statistical mechanics, we briefly address possible generalized constraints under which the present entropy could be extremized, in order to produce canonical-ensemble-like stationary-state distributions for Hamiltonian systems.

Journal Article↗

Ensemble averages and nonextensivity at the edge of chaos of one-dimensional maps.

Ensemble averages of the sensitivity to initial conditions xi(t) and the entropy production per unit of time of a new family of one-dimensional dissipative maps, x(t+1)=1-ae(-1/|x(t)|(z))(z>0), and of the known logisticlike maps, x(t+1)=1-a|x(t)|(z)(z>1), are numerically studied, both for strong (Lyapunov exponent lambda(1)>0) and weak (chaos threshold, i.e., lambda(1)=0) chaotic cases. In all cases we verify the following: (i) both [ln((q)x triple bond (x(1-q)-1)/(1-q); ln((1)x=ln(x] and [S(q) triple bond (1- sigma p(q)(i))/(q-1); S(1)=- sigma p(i)ln(p(i)] linearly increase with time for (and only for) a special value of q, q(av)(sen), and (ii) the slope of and that of coincide, thus interestingly extending the well known Pesin theorem. For strong chaos, q(av)(sen)=1, whereas at the edge of chaos q(av)(sen)(z)<1.

Journal Article↗

Random matrix ensembles from nonextensive entropy.

The classical Gaussian ensembles of random matrices can be constructed by maximizing Boltzmann-Gibbs-Shannon's entropy, S(BGS) = -integral dH[P(H)]ln[P(H)], with suitable constraints. Here, we construct and analyze random-matrix ensembles arising from the generalized entropy S(q) = [1- integral dH [P(H)](q)] /(q-1) (thus, S1 = S(BGS) ). The resulting ensembles are characterized by a parameter q measuring the degree of nonextensivity of the entropic form. Making q-->1 recovers the Gaussian ensembles. If q not equal 1, the joint probability distributions P(H) cannot be factorized, i.e., the matrix elements of H are correlated. In the limit of large matrices two different regimes are observed. When q<1, P(H) has compact support, and the fluctuations tend asymptotically to those of the Gaussian ensembles. Anomalies appear for q>1 : Both P(H) and the marginal distributions P( H(ij) ) show power-law tails. Numerical analyses reveal that the nearest-neighbor spacing distribution is also long-tailed (not Wigner-Dyson) and, after proper scaling, very close to the result for the 2 x 2 case--a generalization of Wigner's surmise. We discuss connections of these "nonextensive" ensembles with other non-Gaussian ones, such as the so-called Lévy ensembles and those arising from soft confinement.

Journal Article↗

Classical infinite-range-interaction Heisenberg ferromagnetic model: metastability and sensitivity to initial conditions.

An N-sized inertial classical Heisenberg ferromagnet, which consists of a modification of the well-known standard model, where the spins are replaced by classical rotators, is studied in the limit of infinite-range interactions. The usual canonical-ensemble mean-field solution of the inertial classical n-vector ferromagnet (for which n=3 recovers the particular Heisenberg model considered herein) is briefly reviewed, showing the well-known second-order phase transition. This Heisenberg model is studied numerically within the microcanonical ensemble through molecular dynamics. In what concerns the caloric curve, it is shown that, far from criticality, the kinetic temperature obtained at the long-time-limit microcanonical-ensemble simulation recovers well the equilibrium canonical-ensemble estimate, whereas, close to criticality, a discrepancy (presumably due to finite-size effects) is found. The time evolution of the kinetic temperature indicates that a basin of attraction exists for the initial conditions for which the system evolves into a metastable state, whose duration diverges as N--> infinity, before attaining the terminal thermal equilibrium. Such a metastable state is observed for a whole range of energies, which starts right below criticality and extends up to very high energies (in fact, the gap between the kinetic temperatures associated with the metastable and the terminal-equilibrium states is expected to disappear only as one approaches infinite energy). To the best our knowledge, this has never before been observed on similar Hamiltonian models, in a noticeable way, for such a large range of energies. For example, for the XY (n=2) version of the present model, such a behavior was observed only near criticality. It is shown also that the (metastable state) maximum Lyapunov exponent decreases with N like lambda(max) approximately N-kappa, where for the initial conditions employed herein (maximal magnetization), kappa=0.225+/-0.030, both above and below the critical point.

Journal Article↗

Constructing a statistical mechanics for Beck-Cohen superstatistics.

The basic aspects of both Boltzmann-Gibbs (BG) and nonextensive statistical mechanics can be seen through three different stages. First, the proposal of an entropic functional (S(BG)=-k Sigma(i)p(i)ln p(i) for the BG formalism) with the appropriate constraints (Sigma(i)p(i)=1 and Sigma(i)p(i)E(i)=U for the BG canonical ensemble). Second, through optimization, the equilibrium or stationary-state distribution (p(i)=e(-betaE(i))/Z(BG) with Z(BG)= Sigma(j)e(-betaE(j)) for BG). Third, the connection to thermodynamics (e.g., F(BG)=-(1/beta)ln Z(BG) and U(BG)=-(partial differential/partial differential beta)ln Z(BG)). Assuming temperature fluctuations, Beck and Cohen recently proposed a generalized Boltzmann factor B(E)= integral (infinity)(0)dbetaf(beta)e(-betaE). This corresponds to the second stage described above. In this paper, we solve the corresponding first stage, i.e., we present an entropic functional and its associated constraints which lead precisely to B(E). We illustrate with all six admissible examples given by Beck and Cohen.

Journal Article↗

Metastability and weak mixing in classical long-range many-rotator systems.

We perform a molecular dynamical study of the isolated d=1 classical Hamiltonian H=1 / 2 summation operator (N)(i=1)L(2)(i)+ summation operator (i not equal j)[1-cos(theta(i)-theta(j))]/r(alpha)(ij); (alpha> or =0), known to exhibit a second order phase transition, being disordered for u identical with U/NN> or =u(c)(alpha,d) and ordered otherwise [U identical with total energy and N identical with (N(1-alpha/d)-alpha/d)/(1-alpha/d)]. We focus on the nonextensive case alpha/d< or =1 and observe that, for u infinity ), where, for all values of alpha/d, kappa(metastable) numerically coincides with one third of its value for u>u(c), hence decreases from 1/9 to zero when alpha/d increases from zero to unity, remaining zero thereafter. This simple connection between anomalies above and below the critical point reinforces the nonextensive universality scenario.

Journal Article↗

Nonequilibrium probabilistic dynamics of the logistic map at the edge of chaos.

We consider nonequilibrium probabilistic dynamics in logisticlike maps x(t+1)=1-a|x(t)|(z), (z>1) at their chaos threshold: We first introduce many initial conditions within one among W>>1 intervals partitioning the phase space and focus on the unique value q(sen)<1 for which the entropic form S(q) identical with (1- summation operator Wp(q)(i))/(q-1) linearly increases with time. We then verify that S(q(sen))(t)-S(q(sen))( infinity ) vanishes like t(-1/[q(rel)(W)-1]) [q(rel)(W)>1]. We finally exhibit a new finite-size scaling, q(rel)( infinity )-q(rel)(W) proportional, variant W(-|q(sen)|). This establishes quantitatively, for the first time, a long pursued relation between sensitivity to the initial conditions and relaxation, concepts which play central roles in nonextensive statistical mechanics.

Journal Article↗

Border between regular and chaotic quantum dynamics.

We identify a border between regular and chaotic quantum dynamics. The border is characterized by a power-law decrease in the overlap between a state evolved under chaotic dynamics and the same state evolved under a slightly perturbed dynamics. For example, the overlap decay for the quantum kicked top is well fitted with [1+(q-1)(t/tau)2](1/(1-q)) (with the nonextensive entropic index q and tau depending on perturbation strength) in the region preceding the emergence of quantum interference effects. This region corresponds to the edge of chaos for the classical map from which the quantum chaotic dynamics is derived.

Journal Article↗

Asymmetric unimodal maps at the edge of chaos.

We numerically investigate the sensitivity to initial conditions of asymmetric unimodal maps x(t+1)=1-a/x(t)/(z(i)) (i=1,2 correspond to x(t)>0 and x(t)<0, respectively, z(i)>1, 0<a< or =2, t=0,1,2,...) at the edge of chaos. We employ three distinct algorithms to characterize the power-law sensitivity to initial conditions at the edge of chaos, namely: direct measure of the divergence of initially nearby trajectories, the computation of the rate of increase of generalized non-extensive entropies S(q), and multi-fractal analysis. The first two methods provide consistent estimates for the exponent governing the power-law sensitivity. In addition to this, we verify that the multi-fractal analysis does not provide precise estimates of the singularity spectrum f(alpha), especially near its extremal points. Such feature prevents to perform a fine check of the accuracy of the scaling relation between f(alpha) and the entropic index q, thus restricting the applicability of the multi-fractal analysis for studying the sensitivity to initial conditions in this class of asymmetric maps.

Journal Article↗