arXiv:2610.08769
Observing the magic Mpemba effect in localized dynamics on a digital quantum computer
Han-Ze Li, Xianquan Yan, Yi-Rui Zhang, Jian-Xin Zhong, Shuo Liu, Ching Hua Lee
Abstract
The Mpemba effect challenges the intuition that states closer to equilibrium must relax faster. We ask whether an analogous magic-ordering reversal can occur in the generation of quantum magic (nonstabilizerness), a key resource for universal quantum computation: can a state with less initial magic overtake one with more and reach its asymptotic value sooner? Here, we uncover interaction-induced dephasing as a distinct mechanism for this magic Mpemba effect, requiring neither transport nor conventional thermalization. For tilted product states in an interacting $\ell$-bit model, a random-phase analysis predicts a reversal of the magic ordering: states with less initial magic attain higher saturated magic, independent of their spatial pattern. Exact simulations of the $\ell$-bit model and localized spin chains further show that these states also saturate earlier. We observe both features on IBM superconducting quantum processors. An effective $\ell$-bit implementation accesses long dephasing times at fixed circuit depth by encoding time in gate angles, while a microscopic spin model exhibits the reversal universally across initial patterns in the localized regime but only selectively in the ergodic regime. These results establish interaction-induced dephasing as a route to anomalous quantum-resource relaxation beyond transport-driven thermalizing dynamics.
AI Summary
Can an initially less magical state generate more magic and reach its own saturation value sooner without transport or thermalization? This paper studies this “heating-side” magic Mpemba effect in interacting localized dynamics, using second stabilizer Rényi entropy $\mathcal M_2$. For tilted product states in a directly encoded $\ell$-bit model, an independent-configuration-phase calculation gives the thermodynamic saturated magic density $\mathcal M_2^{\mathrm{sat}}/N=1-\log_2(1+\cos^4\theta)$. On $\pi/4<\theta<\pi/2$, this increases with tilt while initial magic decreases. The endpoint reversal is independent of the initial binary spatial pattern. Exact $N=14$ simulations find earlier saturation for $\theta=0.50\pi$ than $0.35\pi$ across ferromagnetic, Néel and domain-wall families in interacting $\ell$-bit, random-field, quasiperiodic and Stark-localized models, using a normalized 10% tolerance. IBM hardware measurements resolve magic-ordering crossings in six-qubit interacting $\ell$-bit evolution, implemented exactly at time-independent circuit depth, and in ten-qubit kicked-XXZ target states prepared by independently optimized circuits. Localized kicked dynamics reverses ordering for all three families; the ergodic control does so only for ferromagnetic and domain-wall states. A noninteracting precession control shows no reversal over the measured window. The analytic prediction averages moments before taking logarithms and assumes self-averaging when related to Hamiltonian dynamics; it does not establish saturation times. Hardware resolves crossings, not the full earlier-saturation criterion, and noisy-state diagnostics do not directly certify useful mixed-state magic. The 156-qubit reference reconstructs the phase-model prediction from initial polarizations, without interacting evolution. Finite-size spectral diagnostics do not establish thermodynamic localization.
Central claim may not hold: The abstract claims hardware observation of earlier saturation, although saturation-time inversion is established only numerically and the Discussion explicitly identifies its hardware determination as future work. The abstract describes microscopic reversal as universal across initial patterns, although the microscopic evidence covers only three selected initial-state families and the general pattern-independence argument applies to an effective random-phase ensemble.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- Any commments are welcome!