Resonances, Recurrence Times and Steady States in Monitored Noisy Qubit Systems
S. Ma, S. Tornow, E. Barkai · 2026
Read Resonances, Recurrence Times and Steady States in Monitored Noisy Qubit Systems by S. Ma, S. Tornow, E. Barkai (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-18996.txt.
Introduction
Ma and co-authors study repeat times in noisy qubit systems that are watched step by step. Without noise, repeat times fall on whole numbers, with dips near revival points. Tests on an IBM quantum device show the pattern holds far from revivals. Near revivals, even weak noise breaks it and can turn dips into peaks. Their model pits watch-driven heating against a pull toward a cold state. Sampling time tunes which side wins. Noise and detuning act as linked small factors near resonance.
Catalog record
| CREATOR | S. Ma, S. Tornow, E. Barkai |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 45746 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.18996
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Resonances, Recurrence Times and Steady States in Monitored Noisy Qubit Systems S. Ma, S. Tornow, E. Barkai Abstract We study non-equilibrium steady states and recurrence times in noisy, stroboscopically monitored qubit systems using complete measurements. In the noiseless limit, recurrence times are integer-quantized, with dips to lower integers when sampling approaches revival conditions associated with ergodicity breaking. Using an IBM quantum platform, we find that quantization is robust when sampling far from revivals, but breaks down dramatically near revivals: even weak noise produces large deviations and can invert the expected dips into pronounced peaks. To explain this behavior, we formulate a statistical-physics model of monitored noisy circuits in which monitoring drives an effective infinite-temperature steady state while thermal-like relaxation competes to favor a low-temperature limit. We show that the sampling time tunes a crossover between these regimes, near revivals stabilizing low-temperature behavior, and far from revivals restoring infinite-temperature behavior— with noise strength and detuning acting as coupled small parameters near resonance. I Introduction Classical and quantum recurrence times arise across many disciplines, ranging from chaos theory and the foundations of statistical physics to, more recently, qubit systems and quantum circuits Kac ; Kac1960 ; Poincare ; BocchieriLoinger ; Altman ; Grunbaum2013 ; Grunbaum2019 ; Friedman ; YinLarge ; DidiBarkai ; TornowZiegler ; YinPNAS ; WangEntropy ; Stefanak2026recurrencein ; Nitsche ; Sinkovicz2015 ; Walter2025 ; Sinkovicz ; Karimi . In quantum mechanics, recurrence theory states that an isolated, energy-preserving system with a discrete spectrum returns arbitrarily close to its initial state after sufficiently long times BocchieriLoinger . This result extends the classical Poincaré recurrence theorem Poincare , which guarantees recurrence in finite classical phase spaces. These recurrence theorems are closely related to revival phenomena: under unitary dynamics, there may exist times at which the return probability satisfies |\langle\psi(t)|\psi(0)\rangle|^{2}=1 AverbukhPerelman ; Robinett . In such settings, recurrence is deterministic and was historically invoked to challenge Boltzmann’s ideas of ergodicity and relaxation to steady states Ehrenfest . For macroscopic systems, however, recurrence times of isolated dynamics are typically so long as to be physically irrelevant. In contrast, for small qubit systems, such as those realized on present-day noisy intermediate-scale quantum devices, deterministic recurrence scales can become experimentally accessible, as we demonstrate below. For classical stochastic dynamics Norris ; vanKampen ; Redner2001 , recurrence is naturally formulated in probabilistic terms. For discrete time ergodic Markov processes, Kac’s lemma Kac ; Kac1960 ; Aldous ; Kemeny ; Condamin relates the mean recurrence time to a given state i to its steady state probability, \langle n(i)\rangle=\frac{1}{p_{i}^{\rm ss}}, (1) where p_{i}^{\rm ss} is the steady state probability and \langle n(i)\rangle denotes the mean number of time steps until recurrence. This relation connects a dynamical quantity, the recurrence time, to a static steady state property of the system, provided the dynamics is ergodic. For example, in a random walk on a finite graph, the steady state occupation probabilities alone determine the mean recurrence times to the nodes of the graph. Importantly, in classical systems measurements are typically non-invasive, so that both steady states and mean recurrence times are insensitive to the measurement process. Recently, monitored quantum recurrence has emerged as an active area of research Grunbaum2013 ; Grunbaum2019 ; Friedman ; YinLarge ; DidiBarkai ; TornowZiegler ; YinPNAS ; WangEntropy ; Stefanak2026recurrencein ; Nitsche ; Sinkovicz2015 ; Sinkovicz ; Walter2025 ; Stefanak2026recurrencein ; rbtb-8d27 ; Karimi . In this framework, a quantum system undergoes unitary evolution interrupted by stroboscopic projective measurements Dhar ; VecchioDelVecchio2026 ; ladenburger2025 ; Finocchiaro2025 , which generate a detection record from which recurrence times are defined. Unlike in classical dynamics, measurements fundamentally alter the system’s evolution Skinner ; GoogleQuantumAI2023 ; Koh2023 , rendering recurrence statistics sensitive to the sampling interval and the measurement protocol WisemanMilburn . When the measurements are complete vonNeumann , as defined below, quantum coherences are removed at each measurement, while the recurrence time defined through repeated monitoring remains inherently stochastic Strasberg ; Smirne . Remarkably, in ideal noise-free settings, the interplay between unitary dynamics and monitoring gives rise to quantization of the mean recurrence time: the mean number of measurements \langle n\rangle assumes integer values and undergoes abrupt transitions between them as the sampling time is varied DidiBarkai . Within this ideal theory, sampling at revival times can also lead to pronounced dips to lower integers, reflecting measurement-induced ergodicity breaking tied to the underlying unitary dynamics DidiBarkai ; soldner2025 . The key issue underlying the steady states and recurrence properties of realistic, noisy quantum platforms is the coexistence of two distinct Hamiltonians. The first is generated by a synthetic (engineered) Hamiltonian H_{\rm syn} , implemented through a sequence of quantum gates and governing the intended unitary dynamics. In contrast, thermal relaxation originates from the physical device Hamiltonian H_{\rm phys} , which is fixed by the hardware qubits and their coupling to the environment Martinis2005 ; Koch2007 ; Krantz2019 ; Reimann ; Lvov ; Andersen2025 ; Tuokkola2025 ; Cech ; Hothem2025 . The interplay of these two Hamiltonians, together with repeated measurements, generally drives the system into a non-equilibrium steady state Gotta2026 , defined as the stationary distribution of measurement outcomes, whose properties we aim to characterize. As we show, in realistic noisy qubit platforms our observations cannot be explained by deterministic Poincaré-like revivals, neither by a measurement-free steady-state description, nor by the noise-free quantization picture taken in isolation, even though each remains insightful in its respective limit. Using an IBM quantum processor, we demonstrate that at sampling intervals where the underlying unitary dynamics revives, recurrence is not achieved within one step; instead, it becomes a genuinely stochastic event and can exhibit extreme sensitivity even to weak noise. As a consequence, dips in recurrence times predicted by ideal monitored quantum dynamics Grunbaum2013 ; Friedman ; YinLarge ; DidiBarkai ; TornowZiegler ; YinPNAS ; WangEntropy ; Grunbaum2019 can be strongly suppressed or even inverted into peaks in the mean recurrence time. Moreover, the recurrence statistics need not respect the time reversal symmetry of the ideal theory, consistent with a physical bias toward relaxation into the qubit ground state. Finally, because the act of monitoring imprints an explicit \tau dependence on the effective dynamics, recurrence cannot be inferred from a \tau independent steady state. These observations motivate a theory of monitored recurrence in noisy quantum devices, which we develop in this work to identify noise mechanisms consistent with the data and to reveal how the sampling time controls the non-equilibrium steady state and associated recurrence statistics. II Results II.1 Recurrence times on remote quantum processors In this section, we use remote experiments Georgescu ; Li2023 on IBM quantum processors. The role of these demonstrations is to expose regimes in which existing noise-free recurrence theories (to be explained) break down and to guide the construction of effective theoretical descriptions. In particular, we observe large deviations from the integer mean recurrence time predicted by ideal theory. These deviations are most pronounced near special sampling intervals related to revival times. Idealized noiseless model. We consider a finite-dimensional Hilbert space spanned by qubit states |\psi\rangle in a quantum processor. For a system of N qubits, the Hilbert space dimension is 2^{N} . In the absence of decoherence, the dynamics is generated by a time-reversal-symmetric Hamiltonian H_{{\rm syn}}