arXiv:2609.40270
Testing quantum Gaussianity with constant sample complexity
Mahtab Yaghubi Rad, Ricard Puig, Saksham Hassanandani, Luke Coffman, Jens Eisert, Carlos Bravo-Prieto, Antonio Anna Mele
Abstract
Efficiently testing whether a quantum state possesses a given structure is both a fundamental and a practical task in quantum information. Among the most important structured families are Gaussian states, which underpin quantum optics and many-body physics while defining paradigmatic regimes of efficient classical simulation. Despite the central role of Gaussian states, optimal tests of Gaussianity have remained unknown. Here we give optimal and robust algorithms for testing fermionic and bosonic Gaussian states, with sample complexity independent of system size, particle number, and energy. Remarkably, the protocols implement experimentally feasible measurements associated with known exact symmetry characterizations of Gaussianity, using Bell sampling on two copies for fermions and passive interferometry with photon counting on three copies for bosons. We further obtain tolerant testers requiring joint measurements on a larger, but still constant, number of copies. At the heart of our analysis is a unified approach---which we call the gradient-flow method---that applies across all structured families considered here and underlies our optimal testers: representation theory turns approximate satisfaction of the defining symmetry into quantitative control of the gradient of the rejection probability, allowing a continuous descent from the input state to the target family whose length bounds their distance.
AI Summary
The paper asks whether Gaussianity of an unknown pure quantum state can be tested with sample complexity independent of the number of modes. Existing few-copy symmetry tests characterize Gaussian states exactly, but previous robustness guarantees depended on system size or bosonic energy assumptions. It proves that pure fermionic and bosonic Gaussianity can be tested using $\Theta(\varepsilon^{-2}\log(1/\delta))$ copies, distinguishing Gaussian inputs from inputs at trace distance at least $\varepsilon$ from every pure Gaussian state, with failure probability at most $\delta$. Each round uses two copies and Bell sampling for fermions, or three copies and passive interferometry with photon counting for general bosons. Gaussian inputs are accepted with certainty; other inputs have rejection probability $\Omega(d_{\mathcal G}^2)$, uniformly in mode number. No fermionic parity promise or bosonic energy or moment bound is required. Analogous results cover Slater determinants, centered Gaussian states, and coherent states. The proof converts spectral bounds for overlapping acceptance projectors into a gradient inequality for rejection probability. Gradient flow then reaches an exactly accepted state along a path whose length bounds the input’s distance from the target family. Closest-target expansions supply local sensitivity bounds, while state-discrimination arguments give matching sample lower bounds. Tolerant testing achieves $O((\varepsilon_B-\varepsilon_A)^{-2}\log(1/\delta))$ copies when $\varepsilon_B>(1+\eta)\varepsilon_A$ and $\varepsilon_A\le c\eta$, using joint measurements on $O(1/\eta)$ copies per round. The guarantees assume pure inputs; broader tolerant regimes, extensions beyond purity, and mode-independent testing using only single-copy measurements remain open.
The authors conjectured the main statement underlying the fermionic Gaussianity test and used ChatGPT 5.5 Pro, together with substantial human intuition and mathematical guidance, to develop an initial proof.
Central claim may not hold: The abstract claims Gaussianity testing without a purity qualification, whereas the main guarantees require pure inputs and concern the pure Gaussian family, an essential restriction because testing the full mixed Gaussian family has incompatible lower bounds.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 7 + 77 pages, 2 + 1 figures