Towards robust variational quantum simulation of Lindblad dynamics via stochastic Magnus expansion
Jia-Cheng Huang, Hao-En Li, Yi-Cheng Wang, Guang-Ze Zhang, Jun Li, Han-Shi Hu · 2025
Read Towards robust variational quantum simulation of Lindblad dynamics via stochastic Magnus expansion by Jia-Cheng Huang, Hao-En Li, Yi-Cheng Wang, Guang-Ze Zhang, Jun Li, Han-Shi Hu (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-22099.txt.
Introduction
Jia-Cheng Huang and five co-authors strengthen quantum simulation of open systems in a 2025 paper. Real quantum systems leak into their surroundings, and their decay follows the Lindblad equation. Simulating that leak on a quantum computer, by averaging many jittery paths, is costly and fragile.
Their framework marries a stochastic expansion method to the variational simulation style, letting each path take long, stable strides. Far fewer paths then suffice for an accurate average. Tests on a magnet chain, a light-harvesting plant complex, and a chemistry model held accuracy even where the motion oscillates wildly. The scheme promises steadier simulations for the noisy machines of today.
Catalog record
| CREATOR | Jia-Cheng Huang, Hao-En Li, Yi-Cheng Wang, Guang-Ze Zhang, Jun Li, Han-Shi Hu |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 93 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-22099
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Towards robust variational quantum simulation of Lindblad dynamics via stochastic Magnus expansion Jia-Cheng Huang, Hao-En Li, Yi-Cheng Wang, Guang-Ze Zhang, Jun Li, Han-Shi Hu Abstract Dedicated to the centenary of quantum mechanics. In this paper, we introduce a novel and general framework for the variational quantum simulation of Lindblad equations. Building on the close relationship between the unraveled Lindblad dynamics, stochastic Magnus integrators, and variational quantum simulation, we propose a high-order scheme for solving the quantum state diffusion equation using exponential integrators. This formulation facilitates the simulation of wavefunction trajectories within the established framework of variational quantum algorithms for time evolution. Our algorithm significantly enhances robustness in two key aspects: the stability of the simulation with large time steps, and the reduction in the number of quantum trajectories required to accurately simulate the Lindblad dynamics in terms of the ensemble average. We demonstrate the effectiveness of our algorithm through numerical examples in both classical and quantum implementations, including the transverse-field Ising model (TFIM) with damping, the Fenna–Matthews–Olson (FMO) complex, and the radical pair model (RPM). The simulation accuracy can be systematically improved, and the algorithm remains reliable even in highly oscillatory regimes. These methods are expected to be applicable to a broader class of open quantum systems beyond the specific models considered in this study. I Introduction The dynamics of systems governed by Lindblad equations (or Lindblad master equations, LMEs) [ 1 , 2 , 3 ] give rise to a range of unique physical phenomena, which have attracted considerable theoretical and experimental interest in recent studies. In contrast to unitary dynamics described by the Schrödinger equation, Lindblad dynamics accounts for the interaction between the system and its environment through the introduction of a set of Lindblad jump operators (or bath operators). Due to its versatile modeling capability, Lindblad dynamics can describe a wide range of non-Hermitian quantum behaviors arising in diverse fields. Notably, it can be employed to model processes such as light-harvesting in photosynthesis [ 4 ] , charge and energy transfer in quantum dots and molecules [ 5 ] , and radical pair dynamics in magnetic resonance [ 6 , 7 ] . In addition to its applicability for simulating realistic system-bath interactions, recent advancements have shown that Lindblad dynamics can be a powerful tool for approximation of non-equilibrium dynamics [ 8 , 9 ] , quantum error correction [ 10 , 11 ] , and efficient preparation of quantum Gibbs states and ground states on quantum computers [ 12 , 13 , 14 , 15 ] . These developments have generated significant interest across the fields of quantum information, quantum algorithms, quantum many-body physics, and quantum chemistry. Given the extensive and significant applications of the Lindblad equation, we aim to develop robust numerical methods for simulating Lindblad dynamics, which have the potential to benefit various fields. However, since the many-body density operator has the dimension that scales exponentially with the system size, it can be prohibitively expensive to solve the Lindblad equation using classical numerical methods [ 16 , 17 , 18 , 19 ] . From the perspective of quantum computing, Lindblad evolution describes a completely positive and trace-preserving (CPTP) map, also known as a quantum channel, potentially enabling efficient simulation on quantum computers. While theoretical quantum algorithms that attain near-optimal complexity have been developed [ 20 , 21 , 22 , 23 ] , their implementation is typically much more complicated than that of Hamiltonian simulation [ 24 , 25 ] , mainly owing to the non-Hermitian nature of the problem [ 23 ] . Furthermore, there has been a limited investigation into the potential for developing robust near-term quantum algorithms that can effectively simulate Lindblad dynamics [ 26 ] . Existing near-term approaches include techniques such as unitary decomposition of operators [ 27 , 28 ] , Sz.-Nagy unitary dilation [ 29 , 30 , 31 ] , hybrid quantum channel approach [ 32 , 33 ] , quantum imaginary time evolution (QITE) [ 34 ] , singular value decomposition-based quantum simulation [ 35 , 36 ] , and variational quantum simulation (VQS) of quantum dynamics [ 37 , 38 , 39 ] . Among these, the first five algorithms correspond to either vectorized superoperator formulation or Kraus-form formulation of the Lindblad equation. These two formulations both operate within a Hilbert space of dimensionality 4^{L} , where L is the number of qubits, which corresponds to the size of the many-body density operator ( Fig. 1 (a)). As a result, these methods are computationally expensive and less suitable for near-term quantum devices. In contrast, variational simulation methods, which only require simulating a wavefunction of dimension 2^{L} , are more efficient in this context. The central concept of the variational simulation algorithm for Lindblad dynamics is to unravel the Lindblad equation and reformulate it as a stochastic differential equation (SDE) for the wavefunction [ 16 , 40 , 41 , 42 ] . This formulation primarily includes two algorithms: quantum state diffusion (QSD) and quantum jump (QJ). Following this reformulation, we obtain a Monte Carlo-type wavefunction trajectory algorithm for numerical solution of the Lindblad equation. This idea can also be integrated with variational quantum algorithms to solve the stochastic non-Hermitian dynamics for multiple wavefunction trajectories, and the ensemble average of these pure-state trajectories will converge to the exact many-body density operator. However, due to the variance of the samplings of the random processes, many existing algorithms based on unraveled Lindblad dynamics require a large number of wavefunction samples, extremely small time steps, or other hyperparameters that need careful tuning to achieve a high simulation fidelity. As a result, these algorithms often lack robustness, particularly when high accuracy is required for simulating the Lindblad equation ( Fig. 1 (b)) [ 37 , 38 , 39 ] . In this paper, we introduce a novel implementation of the QSD-based variational simulation. Our algorithm exploits the deep connection among McLachlan’s variational principle for quantum dynamics, the stochastic Magnus expansion for the QSD equation, and exponential integrators for solving SDEs. Specifically, as illustrated in Fig. 1 (c), we derive a new high-order numerical scheme for the linear and nonlinear QSD equations using the stochastic Magnus expansion. A hallmark of such numerical scheme is that it takes the form of an exponential integrator, making the variational simulation feasible ( Fig. 1 (d)). By incorporating numerical schemes of arbitrarily high order, our approach offers a systematic means to enhance the robustness of variational quantum simulation compared to other variational methods, both in terms of ensemble averages and time step management. This enables accurate and stable simulation results at a relatively low cost, while preserving the qubit efficiency inherent to the variational algorithm. Furthermore, the form of Magnus expansion can be explicitly characterized through the evaluation of multiple stochastic integrals, calculated via the Fourier expansion of Brownian bridges, which facilitates the easy derivation of higher-order precision schemes, and the time evolution based on the variational algorithm can be naturally extended to time-dependent scenarios. The remainder of this paper is organized as follows. In Section II , we provide some preliminaries on QSD methods, introduce key notations, and explain the connection between the SDEs and the Lindblad equation. Additionally, we outline the