A No-Go Theorem for Shaping Quantum Resources
Samuel Alperin
TL;DR
The paper proves a rigidity theorem for continuous-variable quantum dynamics, showing that no smooth Hamiltonian can reshape higher-order moments of a Gaussian state without perturbing its mean and covariance. By analyzing the Moyal expansion, it identifies the quadratic (symplectic) subalgebra $ ext{sp}(2N, ext{R})$ as the unique finite-order invariant that preserves the Gaussian moment hierarchy, thereby delineating the continuous-variable analogue of the Gottesman–Knill boundary. Non-quadratic Hamiltonians irrevocably couple Gaussian and non-Gaussian sectors, making universal non-Gaussian resource shaping impossible through smooth Hamiltonian evolution alone. The results have broad implications for CV quantum information, providing analytic diagnostics and clarifying why non-Gaussian resources require measurement, ancilla, or nonlinear interactions beyond quadratic Hamiltonians.
Abstract
The ability to engineer non-Gaussian quantum resources underlies quantum technologies from communication and metrology to universal computation. However, while a number of canonical works have set no-go limits for attaining such resources from Gaussian operations, it is widely assumed that such resources can be tuned freely by non-Gaussian Hamiltonian dynamics. Here we prove a general no-go theorem for such resource shaping: no smooth Hamiltonian dynamics can modify higher-order statistical moments of a continuous-variable state without simultaneously changing its mean and covariance. This analytic constraint implies a rigidity theorem for Hamiltonian quantum control-only quadratic (symplectic) generators preserve the Gaussian moment hierarchy, while every non-quadratic term necessarily couples the Gaussian and non-Gaussian sectors. The theorem identifies the symplectic algebra as the unique invariant subalgebra whose differential representations terminate at finite (second) order within the otherwise infinite Hamiltonian algebra. It thereby defines the analytic boundary between classically simulable Gaussian dynamics and the fully universal non-Gaussian regime-the continuous-variable analogue of the Gottesman-Knill frontier.
