Gaussian quantum Markov semigroups on finitely many modes admitting a normal invariant state
Federico Girotti, Damiano Poletti · 2024
Read Gaussian quantum Markov semigroups on finitely many modes admitting a normal invariant state by Federico Girotti, Damiano Poletti (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-10020.txt.
Introduction
Federico Girotti and Damiano Poletti study Gaussian quantum Markov semigroups in two thousand twenty-four. These maps model how open quantum systems change over time. They are the quantum kin of a classic process for noisy drift. Each one is set by a drift matrix, a spread matrix, and a shift vector.
The pair pin down which of these maps keep a normal steady state. They describe the full set of such states as well. They also study long time acts, like loss of quantum links to the surroundings. Over time the motion tends toward closed dynamics, and they find the speed. Tests of mean returns and return or drift close the work.
Catalog record
| CREATOR | Federico Girotti, Damiano Poletti |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 99 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-10020
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Gaussian quantum Markov semigroups on finitely many modes admitting a normal invariant state Federico Girotti, Damiano Poletti Abstract Gaussian quantum Markov semigroups (GQMSs) are of fundamental importance in modelling the evolution of several quantum systems. Moreover, they represent the noncommutative generalization of classical Orsntein-Uhlenbeck semigroups; analogously to the classical case, GQMSs are uniquely determined by a “drift" matrix 𝐙 \mathbf{Z} and a “diffusion" matrix 𝐂 \mathbf{C} , together with a displacement vector ζ \mathbf{\zeta} . In this work, we completely characterize those GQMSs that admit a normal invariant state and we provide a description of the set of normal invariant states; as a side result, we are able to characterize quadratic Hamiltonians admitting a ground state. Moreover, we study the behavior of such semigroups for long times: firstly, we clarify the relationship between the decoherence-free subalgebra and the spectrum of 𝐙 \mathbf{Z} . Then, we prove that environment-induced decoherence takes place and that the dynamics approaches an Hamiltonian closed evolution for long times; we are also able to determine the speed at which this happens. Finally, we study convergence of ergodic means and recurrence and transience of the semigroup. 1 Introduction Gaussian quantum Markov semigroups (GQMSs) form a class of evolutions on bounded linear operators on the bosonic Fock space \Gamma(\mathbb{C}^{d}) (in this work we will ony deal with finitely many modes) that play a central role in the study of quantum mechanical systems, since they provide convenient models for the descriptions of quantum optical experiments, atomic ensembles, quantum memories and many others; therefore, they have been intensively studied in the field of quantum controlled systems under the name of linear quantum systems (see [ 41 ] and references therein). From a mathematical point of view, they represent the noncommutative counterpart of classical Ornstein-Uhlenbeck semigroups (see [ 10 , 31 , 18 ] ). The name Gaussian comes from the fact that GQMSs can be characterized as the class of semigroups that preserve the set of quantum Gaussian states ( [ 35 ] ): given an initial quantum Gaussian state \rho with mean \mathbf{m}\in\mathbb{R}^{2d} and covariance \mathbf{\Sigma}\in M_{2d}(\mathbb{R}) , its evolution under the semigroup is a family of quantum Gaussian states (\rho_{t})_{t\geq 0} with parameters m_{t} and \mathbf{\Sigma}_{t} satisfying the following Cauchy problem: \displaystyle\frac{d\mathbf{m}_{t}}{dt}=\mathbf{Z}^{*}\mathbf{m}_{t}-\mathbf{\zeta},\quad\mathbf{m_{0}}=\mathbf{m}, (1) \displaystyle\frac{d\mathbf{\Sigma}_{t}}{dt}=\mathbf{Z}^{*}\mathbf{\Sigma}_{t}+\mathbf{\Sigma}_{t}\mathbf{Z}+\mathbf{C},\quad\mathbf{\Sigma}_{0}=\mathbf{\Sigma} for some \mathbf{Z},\mathbf{C}\in M_{2d}(\mathbb{R}) and \mathbf{\zeta}\in\mathbb{R}^{2d} characterizing the semigroup. Moreover, their generator can be formally written in a GKSL form with a quadratic Hamiltonian and linear jump operators, where quadratic and linear are to be understood in terms of creation and annihilation operators. The first step in the study of dynamical systems consists in determining whether they admit any equilibria and characterizing their structure. This was the initial motivation of this article: determine when a GQMS admits a normal invariant state and what is the structure of normal invariant states. One of the main contributions of this work is Theorem 11 , where we provide a complete characterization of those GQMSs admitting a normal invariant state in terms of \mathbf{Z} , \mathbf{C} and \mathbf{\zeta} or, equivalently, in terms of the Hamiltonian and jump operators. Our investigation points out that there are two building blocks: the quantum harmonic oscillator, i.e. the Hamiltonian evolution driven by a multiple of the number operator, and the case when \mathbf{Z} is stable (let us call it a stable GQMS ), i.e. the spectrum of \mathbf{Z} is contained in the half-plane of complex numbers with strictly negative real part. Every GQMS admitting a normal invariant state can be decomposed into the tensor product of several quantum harmonic oscillators with possibly different frequencies and a stable GQMS. Since GQMS are the open quantum systems counterpart of quadratic hamiltonian dynamics, a side result of our investigation is the characterization of quadratic Hamiltonians admitting a ground state (Corollary 12 ). Normal invariant states of the quantum harmonic oscillator are convex combinations of number states; in the case of the tensor product of several quantum harmonic oscillators, this continues to hold unless the frequencies are not \mathbb{N} linear independent: in this case, the only difference is that one needs to include pure states supported on some particular linear combinations of number states and their closed convex hull. On the other hand, in the case of stable \mathbf{Z} , there exists a unique quantum Gaussian invariant state whose mean and covariance are given by the only fixed points of Eq. ( 1 ). Theorem 13 completely characterizes the set of normal invariant states of GQMS showing that they are unitarily equivalent to the tensor product of the unique quantum Gaussian invariant state of the stable part and the normal invariant states of the quantum harmonic oscillators. Interestingly, the family of GQMS admitting a normal invariant state turns out to be well behaved in terms of features of the dynamics for long times. First of all, the properties of the semigroup of being irreducible or possessing a faithful normal invariant state admit easy characterizations in terms of \mathbf{Z} , \mathbf{C} and \mathbf{\zeta} (Corollaries 14 and 15 ). Moreover, the decoherence-free subalgebra \mathcal{N}(\mathcal{T}) , which is the biggest W^{*} -algebra on which the semigroup acts as a group of ∗ -automorphisms (Proposition 3 ), can be characterized in terms of the eigenspaces of \mathbf{Z} corresponding to purely imaginary eigenvalues and can be shown to be a factor of type I (Theorem 5 and Corollary 7 ). The decoherence-free subalgebra is a central object in the study of long-time dynamics of quantum Markov semigroups and its connection to Theorem 11 is the following: it turns out that the whole algebra of bounded operators B(\Gamma(\mathbb{C}^{d})) factorizes as B(\Gamma(\mathbb{C}^{d}))=\mathcal{N}(\mathcal{T})\overline{\otimes}{\cal A} , where {\cal A} is another W^{*} -algebra which is invariant for the semigroup, moreover the action of the semigroup restricted to \mathcal{N}(\mathcal{T}) is unitarily equivalent to the tensor product of several quantum harmonic oscillators, while the dynamics restricted to {\cal A} is unitarily equivalent to a stable GQMS. If we denote by \mathcal{T}:=(\mathcal{T}_{t})_{t\geq 0} the GQMS we are studying, we can prove (Theorem 17 ) that there exists a normal conditional expectation {\cal E}:B(\Gamma(\mathbb{C}^{d}))\rightarrow\mathcal{N}(\mathcal{T})\otimes\mathbb{1} such that for every x\in B(\Gamma(\mathbb{C}^{d})) {\rm w}^{*}\text{-}\lim_{t\rightarrow+\infty}\mathcal{T}_{t}(x)-\mathcal{T}_{t}{\cal E}(x)=0. (2) This shows that for very initial operator x , the dynamics asymptotically approaches the Hamiltonian evolution of {\cal E}(x) driven by several harmonic oscillators. This implies that environment-induced decoherence in the sense of [ 4 ] takes place and the dynamics asymptotically reduces to a closed dynamics on a subsystem of the original space. Making use of an equivalence between a generalized Poincaré inequality and the exponential decay of the semigroup restricted to a suitable subspace ( A ), we can show that the speed at which the convergence in Eq. 2 takes place is completely governed by the spectral gap of the ergodic restriction to {\cal A} ; we recall that the spectral gap of stable GQMS was explicitly found in [ 18 ] . Finally, we are able to describe the convergence of ergodic means and determine the decomposition of the system space into its positive recurrent and transient parts (the null recurrent one turns out to be trivial). We recall that the study of normal invariant states for GQMSs and the convergence of any initial state to equilibrium has already been considered