Data & Code for the article: Causality, localization, and universality of monitored quantum walks with long-range hopping by Sayan Roy, Shamik Gupta, Giovanna Morigi arXiv:2504.12053
A powerful strategy to accelerate quantum-walk-based search algorithms leverages on resetting protocols, where a detector monitors a target site and the evolution of the walker is restarted if no detection occurs within a fixed time interval. The optimal resetting rate can be extracted from the time evolution of the probability S(t) that the detector has not clicked up to time t. We analyze S(t) for a quantum walk on a one-dimensional lattice when the coupling between sites decays algebraically as d^{-α} with the distance d, for α∈(0,∞). At long times, S(t) decays with a universal power-law exponent that is independent of α. At short times, S(t) exhibits a plethora of phase transitions as a function of α. From this, we provide a strategy to determine the optimal resetting rate. We identify two regimes: for α>1, the resetting rate r is bounded from below by the velocity with which information propagates causally across the lattice; for α<1, instead, the long-range hopping tends to localize the walker: The optimal resetting rate depends on the size of the lattice and diverges as α→0. Our strategy directly connects local measurement outcomes with the global dynamics encoded in S(t). We derive simple models explaining our numerical results, shedding light on the interplay of long-range coherent dynamics, symmetries, and local quantum measurement processes in determining equilibrium. Our findings offer experimentally testable predictions and provide new physical insights on optimizing quantum search through resetting.
Optimal spatial searches with long-range tunnelingby Emma C. King, Moritz Linnebacher, Peter P. Orth, Matteo Rizzi, Giovanna Morigi [arXiv:2501.08148] This Python notebook computes and visualizes the data for χα, cf. Fig. 3 of the manuscript, for (hypercubic) lattices with dimensions d=1,2,3 and α∈[0,10]. The implementation is sufficiently general such that results for other dimensions, different p-norm distance metrics, and various ranges of α may be computed with ease. The repository contains the precomputed data for parameter regimes that include those reported in the manuscript. All computations are based on analytic expressions for the Laplacian eigenvalues Eα(k) and the summations Sℓ(α), which are provided in the manuscript's main text. - - - - Repository Contents[24.01.2025] Open-source - Optimal spatial searches.ipynb: Main notebook containing all calculations and visualizations.chi_dim1.json, chi_dim2.json, chi_dim3.json: Precomputed data for χα values using the Euclidean norm.chi_dim1_MH.json, chi_dim2_MH.json, chi_dim3_MH.json: Precomputed data for χα values using the Manhattan norm.chi_density_plot_Euclidean.png and chi_density_plot_Manhattan.png: Output density contour plots. Relevant PublicationThis repository accompanies an article preprint, please see:- Optimal spatial searches with long-range tunneling- https://arxiv.org/abs/2501.08148 InstallationThis code is tested with Python 3.11.5 and requires the following libraries:- numpy- matplotlib- json- tqdm- itertools- scipy- multiprocessing ContactFor any queries or issues, please contact Emma King at emma.king@physik.uni-saarland.de.
In a quantum walk, the first-detection return probability (FDRP) characterizes salient features, determining whether the quantum walk is transient or recurrent. We study the FDRP of quantum walks on a chain where the initial site is stroboscopically monitored by a detector and the walker performs long-range hopping between sites. We assume that the hopping strength decays with the distance d as d−α and α≥0 and show that the power-law exponent α critically determines the behavior of the FDRP. The value α=1 separates recurrent (α<1) from transient (α>1) quantum walks through a continuous phase transition in the total detection probability. For α<1, strong long-range hopping induces localization, resulting in unit total detection probability. Instead, for α>1 the long-range walk is transient and the return probability decays algebraically as a function of time as t−β. The associated decay exponent β features nonanalytic points as a function of α. Such singularities are not exclusively determined by the low-energy spectrum, but are caused by the interference between infrared and ultraviolet energy modes induced by projective measurements, signalling the emergence of critical behavior intrinsic to the non-unitary dynamics. These dynamics are solely controlled by tuning the long-range exponent α and can thus be experimentally probed in atomic and molecular systems.