arXiv:2610.12929
An Instability of the Quantum Markov Length
David C. Newsom, Michael Levin
Abstract
The Markov length, the length scale governing the exponential decay of conditional mutual information (CMI) in a quantum many-body state, has proven useful for characterizing mixed-state phases of matter. Nevertheless, we show that a state's Markov length is not generally stable under weak local channels. The failure of stability occurs already at a pure product state. We demonstrate this explicitly in a one-dimensional fermionic chain, where a fine-tuned depth-two circuit of nearest-neighbor Gaussian channels maps the vacuum to a state whose CMI decays \emph{algebraically} and whose Markov length is therefore infinite, even when the channels are taken arbitrarily close to the identity. The algebraic decay is rooted in the nonanalyticity of the von Neumann entropy at rank-deficient density operators. We exhibit an analogous instability on an ordinary qubit chain. More generally, we show that the Markov length is unstable in any translation-invariant state with even a single rank-deficient reduced density operator. These results pose a challenge to recent definitions of mixed-state phase equivalence based on the Markov length.
AI Summary
The paper asks whether finite quantum Markov length—the length scale for exponential decay of conditional mutual information (CMI)—is stable under sufficiently weak local channels. It gives counterexamples starting from a pure product state, with implications for mixed-state phase definitions based on bounded Markov length or local reversibility. The main construction is a depth-two circuit of nearest-neighbor Gaussian channels on a one-dimensional fermionic vacuum. Each channel approaches the identity as its strength $\epsilon\to0$, yet the output has infinite Markov length for every $0<\epsilon<\pi/2$. A dark mode produces a quadratic zero in the correlation-matrix spectrum. Combined with the nonanalyticity of von Neumann entropy at zero eigenvalues, this yields interval entropy $S_N=N\bar{s}+\Delta-c/N^2+O(N^{-3})$, where $N$ counts two-site cells and $c=\zeta(3)\sin^2\epsilon/[16(1+\cos^2\epsilon)]$. Consequently, CMI decays algebraically, as $N^{-2}$ when region sizes scale together. The derivation uses explicit spectral quantization and singular Euler–Maclaurin asymptotics. Extensions include local Lindbladian evolution, canonical bosons, an explicit local qubit circuit, and translation-invariant states having any rank-deficient finite-region reduction, using an extracted pure virtual-qubit layer. Integer-order Rényi entropy combinations instead decay exponentially. The instability requires fine-tuning: changing channel weights or giving the initial product state nonzero temperature restores exponential decay in the examples. Stability for states with all finite-region reductions full rank, including finite-temperature Gibbs states generally, remains unresolved, as does the corresponding classical question.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 5 Pages + Appendices, 3 Figures, Minor adjustments