We show that the dynamics of generic quantum systems concentrate around their equilibrium value when measuring at arbitrary times. This means that the probability of finding such values away from that equilibrium is exponentially suppressed, with a decay rate given by the effective dimension. Our result allows us to place a lower bound on the recurrence time of quantum systems, since recurrences corresponds to the rare events of finding a state away from equilibrium. In many-body systems, this bound is doubly exponential in system size. We also show corresponding results for free fermions, which display a weaker concentration and earlier recurrences.
We show that the dynamics of generic quantum systems concentrate around their equilibrium value when measuring at arbitrary times. This means that the probability of finding such values away from that equilibrium is exponentially suppressed, with a decay rate given by the effective dimension. Our result allows us to place a lower bound on the recurrence time of quantum systems, since recurrences corresponds to the rare events of finding a state away from equilibrium. In many-body systems, this bound is doubly exponential in system size. We also show corresponding results for free fermions, which display a weaker concentration and earlier recurrences.
Generic quantum systems --as much as their classical counterparts-- pass arbitrarily close to their initial state after sufficiently long time. Here we provide an essentially exact computation of such recurrence times for generic non-integrable quantum models. The result is a universal function which depends on just two parameters, an energy scale and the effective dimension of the system. As a by-product we prove that the density of orthogonalization times is zero if at least nine levels are populated and connections with the quantum speed limit are discussed. We also extend our results to integrable, quasi-free fermions. For generic systems the recurrence time is generally doubly exponential in the system volume whereas for the integrable case the dependence is only exponential. The recurrence time can be decreased by several orders of magnitude by performing a small quench close to a quantum critical point. This setup may lead to the experimental observation of such \emph{fast} recurrences.
One of the main questions of research on quantum many-body systems following unitary out of equilibrium dynamics is to find out how local expectation values equilibrate in time. For non-interacting models, this question is rather well understood. However, the best known bounds for general quantum systems are vastly crude, scaling unfavorable with the system size. Nevertheless, empirical and numerical evidence suggests that for generic interacting many-body systems, generic local observables, and sufficiently well-behaved states, the equilibration time does not depend strongly on the system size, but only the precision with which this occurs does. In this discussion paper, we aim at giving very simple and plausible arguments for why this happens. While our discussion does not yield rigorous results about equilibration time scales, we believe that it helps to clarify the essential underlying mechanisms, the intuition and important figure of merits behind equilibration. We then connect our arguments to common assumptions and numerical results in the field of equilibration and thermalization of closed quantum systems, such as the eigenstate thermalization hypothesis as well as rigorous results on interacting quantum many-body systems. Finally, we complement our discussions with numerical results - both in the case of examples and counter-examples of equilibrating systems.
We discuss the local equilibration of closed systems using the relative purity, a paradigmatic information-theoretic distinguishability measure that finds applications ranging from quantum metrology to quantum speed limits. First we obtain an upper bound on the average size of the fluctuations on the relative purity: it depends on the effective dimension resembling the bound obtained with the trace distance. Second, we investigate the dynamics of relative purity and its rate of change as a probe of the speed of fluctuations around the equilibrium. In turn, such speed captures the notion of how fast some nonequilibrium state approaches the steady state under the local nonunitary dynamics, somehow giving the information of the quantum speed limit towards the equilibration. We show that the size of fluctuations depends on the quantum coherences of the initial state with respect to the eigenbasis of the Hamiltonian, also addressing the role played by the correlations between system and reservoir into the averaged speed. Finally, we have derived a family of lower bounds on the time of evolution between these states, thus obtaining an estimate for the equilibration time at the local level. These results could be of interest to the subjects of equilibration, quantum speed limits, and also quantum metrology.
The evolution of an isolated quantum system inevitably exhibits recurrence: the state returns to the vicinity of its initial condition after finite time. Despite its fundamental nature, a rigorous quantitative understanding of recurrence has been lacking. We establish upper bounds on the recurrence time, trec≲texit(ε)(1/ε)d, where d is the Hilbert-space dimension, ε the neighborhood size, and texit(ε) the escape time from this neighborhood. For pure states evolving under a Hamiltonian H, estimating texit is equivalent to an inverse quantum speed limit problem: finding upper bounds on the time a time-evolved state ψt needs to depart from the ε-vicinity of the initial state ψ0. We provide a partial solution, showing that under mild assumptions texit(ε)≈ε/Δ(H2), with Δ(H2) the Hamiltonian variance in ψ0. We show that our upper bound on trec is generically saturated for random Hamiltonians. Finally, we analyze the impact of coherence of the initial state in the eigenbasis of H on recurrence behavior.
There has been considerable recent progress in clarifying the conditions under which quantum systems equilibrate. We demonstrate that effective equilibration of an isolated system can occur under much weaker conditions than previously considered, by requiring only an assumption of sufficiently complex dynamics (which encodes a condition on the complexity of the system eigenvectors) and natural information-theoretic constraints resulting from the infeasibility of collecting an astronomically large amount of measurement data. In particular, we prove that, for generic Hamiltonian systems, the measurement statistics for maximally fine-grained measurements on pure states are effectively indistinguishable from the micro-canonical distribution after a finite equilibration time. We demonstrate that our assumptions correspond to physically realistic conditions by numerically confirming their validity for both generic many-body systems limited to two-body interactions and the quantum kicked-top in a globally chaotic regime.
Recurrence time is evaluated for some initial quantum states in the one-dimensional Bose gas with repulsive short-range interactions. In the relatively strong and weak coupling cases some different types of initial states show almost complete recurrence and the estimates of recurrence time are proportional to some powers of the system size at least in some range of the system size. They are much longer than in the case of free particles such as 100 times. In the free-bosonic and free-fermionic regimes we evaluate the recurrence time rigorously, which is proportional to the square of the system size. The estimate of recurrence time is given by the order of ten milliseconds in the corresponding experimental systems of cold atoms trapped in one dimension of ten micrometers in length. It is much shorter than the estimate in a generic quantum many-body system, which may be as long as the age of the universe.