arXiv:2610.06736
Characterizing Photonic Partial Distinguishability Through Gram Matrix Tomography
Marcin Kotowski, Marco Robbio, Oliver Reardon-Smith, Benoit Seron, Leonardo Novo, Ernesto F. Galvão, Michał Oszmaniec
Abstract
Partial distinguishability affects multiphoton interference, which is central to large-scale photonic experiments such as Boson Sampling and to applications of multiphoton states across quantum technologies. Characterizing the internal degrees of freedom responsible for this behavior is challenging: in many scenarios they cannot be accessed directly and even their dimension or physical nature may be unknown \emph{a priori}. Here we introduce Gram-matrix tomography for independently prepared photons with pure or nearly pure internal states. For pure states, the Gram matrix contains all information required to predict the statistics of experiments that are insensitive to the internal degrees of freedom. We reconstruct this matrix, up to physically irrelevant phase choices, from a quadratic number of measurable Bargmann invariants (quantities encoding overlap magnitudes and collective phases). We access these quantities using two complementary approaches: targeted Fourier interferometers and two- and three-body photon-number correlations measured in a single random interferometer. We derive explicit error bounds that control the total-variation distance between predicted and actual outcome distributions for any subsequent experiment that is insensitive to the internal states. These guarantees account for finite sampling and deviations from purity. Crucially, the total sample complexity of the protocol is independent of the internal Hilbert-space dimension and polynomial under mild assumptions on overlaps of internal states. Our protocol uses standard interferometric measurements and turns them into a robust and operationally grounded characterization of partially distinguishable photons.
AI Summary
The paper asks how to characterize photonic partial distinguishability when internal degrees of freedom cannot be resolved and their dimension is unknown. For independently prepared photons with pure internal states, it targets the $n\times n$ overlap Gram matrix, modulo individual-state phases. This matrix determines the external state and therefore all statistics of measurements insensitive to internal degrees of freedom. The reconstruction uses $O(n^2)$ Bargmann invariants: pairwise overlaps determine magnitudes, while cycles selected through a spanning tree determine phases. Noisy estimates are projected onto positive-semidefinite matrices with unit diagonal. Two measurement routes are developed: dedicated Fourier interferometers for individual invariants, and linear inversion of second- and third-order photon-number correlations from one random interferometer. The latter suffices when a reference photon overlaps with every other photon. The main operational bound is $\mathrm{TVD}(p_G,p_{\widehat G})\leq\frac{\sqrt n}{2}\|G-\widehat G\|_F$, uniformly over external measurements. With relevant invariant magnitudes bounded below by constants, Fourier measurements require $O(n^4/\delta^2)$ runs for Frobenius error $\delta$, or $O(n^5/\delta^2)$ for external-state trace-norm error $\delta$. A packing argument gives a worst-case lower bound $\Omega(n^2/\delta^2)$, including collective external measurements. Bounds separate sampling error from impurity bias for nearly pure product inputs. Efficiency depends on overlap structure; small cycle invariants can cause exponential sampling costs. Random-interferometer conditioning is proved only for second-order inversion in a dilute regime; broader conditioning claims rely on small-system numerics. Simulations find larger errors for the single-interferometer route at equal shot budgets. Loss, calibration errors, correlated preparations, and general mixed-state tomography remain outside the guarantees.
ChatGPT 5.6 Sol was used for auxilliary computations in Appendix C and D. It also provided the proof of \Cref{prop:tvd-continuity} and \Cref{th:sample-lower-bound}.
Error found: The printed Gram-matrix estimator uses the cycle phase rather than its conjugate, so it can return the wrong Gram matrix and external state even with exact data; reversing the phase orientation repairs this local error. The main-text discussion presents high-probability conditioning with linearly many interferometer modes more strongly than established: the rigorous theorem covers only second-order inversion in a substantially more dilute regime, while third-order and linear-mode conditioning are numerical evidence.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.