Space-time generalization of mutual information

Paolo Glorioso,Xiao-Liang Qi,Zhenbin Yang
2024-01-05
Abstract:The mutual information characterizes correlations between spatially separated regions of a system. Yet, in experiments we often measure dynamical correlations, which involve probing operators that are also separated in time. Here, we introduce a space-time generalization of mutual information which, by construction, satisfies several natural properties of the mutual information and at the same time characterizes correlations across subsystems that are separated in time. In particular, this quantity, that we call the \emph{space-time mutual information}, bounds all dynamical correlations. We construct this quantity based on the idea of the quantum hypothesis testing. As a by-product, our definition provides a transparent interpretation in terms of an experimentally accessible setup. We draw connections with other notions in quantum information theory, such as quantum channel discrimination. Finally, we study the behavior of the space-time mutual information in several settings and contrast its long-time behavior in many-body localizing and thermalizing systems.
Quantum Physics,High Energy Physics - Theory
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the problem of correlation measurement between two space - time regions in quantum information theory. Specifically, the traditional Mutual Information (MI) can well describe the correlation between two spatially separated subsystems, but when it comes to temporal separation, the traditional definition of Mutual Information is no longer applicable. This is because, in quantum field theory, for time - like separated regions, it is usually impossible to define a joint state, especially when region B is within the future light cone of region A. Therefore, this paper aims to introduce a new measurement - Space - Time Mutual Information (STMI) - to quantify the correlation between two arbitrary space - time regions A and B, even if these regions are temporally separated. ### Main contributions 1. **Define STMI**: Based on the idea of hypothesis testing, the author proposes the concept of STMI. In the hypothesis experiment, researchers can detect the correlation between these regions by coupling an ancilla to specific regions in the main system. STMI is defined by optimizing the relative entropy under different coupling schemes. 2. **Property analysis**: The paper explores some basic properties of STMI. For example, when A and B are spatially separated, STMI degenerates into the traditional Mutual Information; STMI can serve as an upper bound of the dynamic correlation function; in some cases, STMI satisfies the Markov property in space - time. 3. **Applications and examples**: The author studies the behavior of STMI in different scenarios, especially comparing the long - time behaviors of the Many - Body Localization (MBL) system and the thermalizing system. ### Formula analysis - **Definition of STMI**: \[ J_N(A:B)=\frac{1}{N}\sup_{V_A, V_B}S(\sigma_W^{(N),1}\|\sigma_W^{(N),0}) \] where $\sigma_W^{(N),1}$ and $\sigma_W^{(N),0}$ respectively represent the final states of the auxiliary system W in two different situations, and $S(\cdot\|\cdot)$ represents the relative entropy. - **Simplified form of STMI**: \[ J_N(A:B)=\frac{1}{N}\sup_V S(\rho_{BNW}\|\rho_B^{\otimes N,0}\otimes\rho_W) \] Here, $\rho_{BNW}$ is the joint state after coupling, $\rho_B^{\otimes N,0}$ is the evolution state of the unperturbed B subsystem, and $\rho_W$ is the state of the auxiliary system W. - **Properties of STMI**: - **Monotonicity**: STMI is non - increasing when a quantum channel is applied to A or B alone. - **Upper bound**: STMI can serve as an upper bound of the dynamic correlation function, for example: \[ J_N(A:B)\geq\frac{1}{8}\left(\frac{-i\text{Tr}(\rho_{\text{in}}[O_B(t),O_A])}{\|O_A\|_\infty\|O_B\|_\infty}\right)^2 \] \[ J_N(A:B)\geq\frac{1}{8}\left(\frac{\text{Tr}(\rho_{\text{in}}\{O_B(t),O_A\})_c}{\|O_A\|_\infty\|O_B\|_\infty}\right)^2 \] ### Conclusion By introducing STMI, the paper...