arXiv:2610.06769
Fast convergence of propagation algorithms for open quantum systems
Giorgio Facelli, Hamza Fawzi, Omar Fawzi, Cambyse Rouzé, Sam Slezak, Daniel Stilck França
Abstract
We establish general conditions for the rapid convergence of propagation algorithms for computing expectation values of local observables in quantum many-body systems, covering both spin and fermionic systems. Our results apply to evolutions that are sufficiently contractive, providing a general mechanism by which contractivity controls the growth generated by local interactions and enables efficient classical simulation. As a first application, we consider noisy time evolution generated by local Hamiltonians on arbitrary interaction graphs. We show that the dynamics can be efficiently simulated when the noise strength $λ$ is sufficiently large compared to the degree of the interaction graph. More generally, for systems with an interaction strength $u$, our bounds determine a time horizon $t_{\text{max}}$ as a function of the interaction strength $u$, the noise strength $λ$, and the interaction structure below which the algorithm is efficient. In the noiseless limit, $λ=0$, our analysis extends the efficiently simulable time scale from $\log(1/u)$ proved in Facelli, Fawzi, and Fawzi (2026) to $t_{\text{max}} \sim 1/u$ matching the recent improvement by Zhao, Marvian and Tong (2026). As a second application, we apply the same framework to the computation of local observables in Gibbs states of local Hamiltonians. We consider a pseudo-Lindbladian whose steady state is the Gibbs state and show that, at sufficiently high temperature, its propagation algorithm converges rapidly. In particular, for weakly-interacting systems with interaction strength $u$, our result applies up to inverse temperatures $β\sim\log(1/u)$.
AI Summary
The paper asks when degree-truncated Pauli/Majorana propagation can rigorously approximate local-observable dynamics and thermal expectations. It develops a contraction-based analysis: exponential weights on string degree control high-degree tails, and stability plus short-step error bounds control accumulated propagation error. For dynamics generated by $\mathcal L=i[H_0+uV,\cdot]-\lambda\mathcal N$, $H_0$ preserves degree, interactions increase it by at most $\delta$, and the Hamiltonian is $\Delta$-sparse. A time-dependent weighted Frobenius norm contracts whenever $a'(t)\leq\lambda-2\Delta u\sinh(\delta a(t))$. This yields an all-time guarantee for $\lambda\geq2\Delta u\sinh(1)$ and a finite simulation window below that threshold, scaling as $1/(\Delta u)$ without noise. For fixed-degree observables, cutoff $\ell=d+O(\log((1+t)/\varepsilon))$ gives normalized Frobenius error $\varepsilon\|A\|_F$ with runtime $N^{O(\log((1+t)/\varepsilon))}$. This norm controls ensemble-averaged expectation errors, not arbitrary-state worst-case errors. For fermionic Gibbs expectations, an auxiliary, generally non-completely-positive pseudo-Lindbladian fixes the Gibbs state in the dual picture. Weighted coefficient $\ell_1$ contraction proves relaxation to the thermal expectation times identity; a Dyson expansion controls interactions and supplies computable short steps. With geometrically local quadratic hopping and sparse quartic interactions, a sufficient condition is $u\Delta e^{C\beta}(e^{2\beta}-1)\leq c$, permitting $\beta=O(\log(1/u))$ at fixed degree. The deterministic runtime is $(N+1)^{O(d+\log(1/\varepsilon))}\operatorname{poly}(d,1/\varepsilon)$. The paper notes weaker runtime and temperature guarantees than existing randomized cumulant-expansion methods; improved selective truncation is left open.
OpenAI's ChatGPT Sol 5.6 and ChatGPT Astra 6 assisted with exploring and checking proof strategies and technical details.
Error found: The formal dynamics theorems omit a coefficient bound on V independent of u, making their weighted-contraction conclusion false under the written assumptions, although the introduction states the intended missing restriction. The introductory thermal theorem advertises a condition uniform in interaction degree Δ that is not established by the formal contraction theorem's exponentially stronger dependence on Δβ.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 31 pages