arXiv:2610.26222
A Readout-Aware Theory of Approximate Heisenberg Propagation
Yubraj Bhandari, Jakob S. Kottmann, Adelina Bärligea
Abstract
Classical methods for computing quantum expectation values often propagate observables in the Heisenberg picture. To remain tractable, these methods rely on approximations that discard operator components during evolution. Yet the magnitude of these components alone does not determine the resulting expectation-value error, which also depends on their subsequent evolution and the input state. We develop \emph{generalized Pauli propagation}, a common algebraic framework encompassing Pauli propagation, Lie-algebraic simulation, and Krylov subspace methods. Using weighted commutator graphs, we derive exact identities that relate approximation errors to the signed contributions of omitted and discarded operator components. Our central result establishes \emph{delayed observability}, identifying conditions under which these contributions vanish at low orders in propagation time. We use this structure to develop finite-order error estimates and state-dependent selection rules. We further prove that retaining the symmetry-preserving part of a weak perturbation eliminates first-order expectation-value errors under suitable symmetry conditions. By revealing how approximation errors affect the desired expectation value, our framework provides a basis for more informative error analysis and the development of readout-aware simulation strategies.
AI Summary
Approximate Heisenberg propagation discards operator components, but their norms alone do not determine the error in a chosen expectation value. This paper formulates Pauli, Lie-algebraic, symmetry-adapted, and fixed-generator Krylov propagation as the bilinear transfer problem $w^\top e^{tA}s$, with a weighted commutator graph connecting observable coefficients to state-dependent readout coordinates. Exact continuous and discrete defect-transfer identities express expectation-value errors as signed contributions from initially omitted components and residuals generated during approximate evolution. The central delayed-observability theorem assumes that positive-grade operators have zero input-state expectation and each commutator lowers grade by at most $\delta$. A defect of minimum grade $g$ then satisfies $r^\top A^m s=0$ for $m<\lceil g/\delta\rceil$. This follows by bounding grade changes along graph walks. Weight-space selection rules extend the construction to Cartan-diagonal and generalized coherent-state inputs. Truncated transfer expansions provide signed error estimates with exponential-series remainder bounds. A separate perturbative result gives $O(\varepsilon^2)$ readout error when source and readout lie in an unperturbed invariant sector and the approximation retains the within-sector perturbation. Small-system numerics test these statements. In a six-qubit Ising circuit, eight additional transfer orders reduce estimation discrepancy to approximately $2.1\times10^{-12}$ for an actual truncation error of $4.2\times10^{-5}$. However, deeper random circuits yield median relative estimation error above one. Including estimator cost, correction beats tested uncorrected alternatives at comparable generated work in only about 21% of Ising and 15% of circuit comparisons. Delayed observability alone guarantees neither efficient propagation nor accurate fixed-order estimates.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 14 pages main text, 13 pages appendices, 6 figures, 1 table