arXiv:2610.20712
Surviving readout errors and incomplete postselection in monitored quantum circuits
Murtaza Ali Mir, Teemu Ojanen, Ali G. Moghaddam
Abstract
We study how readout errors affect dynamics in monitored quantum circuits. Such errors make postselection effectively incomplete, conditioning trajectories on imperfect rather than true records. Focusing on U(1)-symmetric circuits, we find that the charge-sharpening transition survives at a finite error rate, with the sharp phase's vanishing fluctuations replaced by a finite steady-state plateau. The charge fluctuations show a shallow minimum near the ideal transition point as a function of measurement rate, signaling the competition between measurement-induced purification and noise-induced mixing. By decomposing the fluctuations into quantum and classical contributions, we show the minimum marks where incomplete postselection still retains enough trajectory memory to attenuate noise-induced mixing, leaving a detectable signature of the transition even in a noisy setting. Mid-circuit measurements are far harder to mitigate against readout errors than standard end-of-circuit ones, because the errors corrupt the postselection itself; the decomposition therefore offers a minimal mitigation framework and a route to locate the transition directly from experimental data. Our results show that charge sharpening survives readout errors in a quantitative and experimentally accessible way.
AI Summary
The paper asks how imperfect mid-circuit measurement records affect charge sharpening in $\mathrm{U}(1)$-symmetric monitored circuits. Independent assignment errors make conditioning on reported outcomes a mixture over true trajectories. It studies charge-conserving Haar-random two-qubit circuits initialized in $|+\rangle^{\otimes L}$, using stochastic pure-state simulations to estimate conditional mixed-state charge variances. The main numerical results, predominantly for $L=12$ and times up to $T=8L$, show that readout noise replaces vanishing charge fluctuations with a finite plateau while retaining a distinction between slow relaxation at weak monitoring and rapid relaxation at strong monitoring. For weak-to-moderate symmetric error rates, approximately $\eta\lesssim0.1$, the normalized variance develops a shallow minimum near the ideal sharpening threshold $p_\#\sim0.1$, followed by an approximately measurement-rate-independent plateau for $p_m\gtrsim0.4$. At completely random readout, $\eta=1/2$, conditioning provides no information and the normalized variance is one. An exact total-variance decomposition separates the noise-free trajectory-averaged variance from an additional variance of trajectory charge means. An approximate noise factor $4\eta(1-\eta)$ describes the latter well at strong monitoring; intermediate-rate deviations are attributed to correlations retained through trajectory history. Subtracting this approximate contribution is proposed as a diagnostic of the underlying transition. The evidence concerns small systems and finite observation times, rather than thermodynamic critical scaling. The classical contribution is not directly easy to measure, and the proposed subtraction requires theoretical input. Accounting for intermediate-rate memory effects and developing general mitigation of corrupted mid-circuit records remain open.
Central claim may not hold: The small-system, finite-time numerical evidence does not establish the claimed survival of a finite-noise charge-sharpening phase transition. At full monitoring and symmetric readout error below one half, the complete accumulated record determines the conserved charge increasingly accurately, forcing the average conditional charge variance to vanish rather than approach the claimed nonzero steady-state plateau. The proposed experimental diagnostic requires an approximate theoretical subtraction and is not demonstrated to locate a finite-noise transition directly from experimental data. The trajectory algorithm as described omits likelihood reweighting between ensemble components and therefore generally computes the wrong state conditioned on the noisy measurement record.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 8 pages, 4 figures