arXiv:2610.01163
Robustness hierarchy of bipartite quantum correlations under noisy dynamics
Shakib Daryanoosh
Abstract
Quantum correlations are fragile under noise, motivating quantitative measures that track not only their presence but also their degradation during physical evolution. We develop a robustness framework for the hierarchy of Bell nonlocality, EPR steering, and entanglement, establishing an ordering of the corresponding robustness measures and relating robustness to noise-induced transition times. We further formulate the instantaneous robustness decay along Lindblad dynamics as a directional derivative determined by both the evolution and the robustness functional. Applications to isotropic depolarizing noise and amplitude damping illustrate, respectively, the resulting resource transition times and the directional dependence of steering robustness. The framework provides an approach to characterizing the degradation and lifetime of bipartite quantum correlations under noisy dynamics.
AI Summary
The paper studies how robustness measures for Bell nonlocality, EPR steering, and entanglement evolve under bipartite noisy dynamics. It defines $R_{\mathfrak C}(\rho)$ as the minimum mixing weight of a state in a convex classical set $\mathfrak C$ needed to bring $\rho$ into that set. Feasible-set inclusion gives $R_{\mathrm{LHV}}\leq R_{\mathrm{LHS}}\leq R_{\mathrm{SEP}}$, and applying a classical-set-preserving channel to an optimal decomposition proves monotonicity. For fixed-reference mixing, the threshold noise fraction is $R/(1+R)$. Along differentiable Lindblad trajectories, the instantaneous change is expressed by the chain rule, $\dot R=DR(\rho)[\mathcal L(\rho)]$, with a Hilbert–Schmidt Cauchy–Schwarz bound. For depolarizing Werner states, the transition time is $t_{\mathfrak C}=\gamma^{-1}\ln[(1-\wp)/(1-\wp_{\mathfrak C})]$. The paper uses thresholds $\wp=1-1/\sqrt2$ and $2/3$ for Bell correlations and separability, respectively. The former is specifically the CHSH threshold, not a general Bell-locality boundary. Its claimed identification of absolute robustness with white-noise robustness also restricts the twirled reference family to $\wp\leq1$, excluding valid Werner reference states beyond that range. For a singlet under one-sided amplitude damping, PPT establishes entanglement at every finite time, while CHSH violation ends at $t=\gamma_{\rm a}^{-1}\ln2$. Direction-dependent steering profiles are formulated but not evaluated. Stated limitations include measurement-scenario-dependent steering thresholds, unresolved robustness optimizations for general trajectories, and the differentiability assumption; nonsmooth derivatives, non-Markovian dynamics, correlated noise, and higher dimensions are left for future work.
Central claim may not hold: The Werner application incorrectly equates absolute robustness with white-noise robustness by excluding valid separable reference states with Werner parameter above 1, producing incorrect robustness values and decay rates. The Werner application labels the CHSH threshold as the exact full Bell-locality boundary, invalidating its claimed exact LHV transition time. The amplitude-damping application establishes only loss of CHSH violation at η=1/2, not entry into the LHV set used by its robustness framework. The amplitude-damping application claims to illustrate directional steering robustness without evaluating either directional robustness profile or steering threshold. The main-text amplitude-damped state omits two diagonal terms and is not normalized, although the appendix supplies the correct state.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 12 pages, 1 figure