arXiv:2610.12070
A deterministic polynomial chaos Galerkin method for non-Markovian quantum state diffusion
Quanhui Zhu, Zhenning Cai
Abstract
Developing general-purpose numerical methods for non-Markovian quantum state diffusion has long been a challenge in the simulation of open quantum systems. This paper proposes a deterministic polynomial chaos Galerkin method, which is easy to implement, accurate, and efficient. By combining a low-rank decomposition of bath correlations with Galerkin projection onto a polynomial chaos basis, the method yields a deterministic formulation compatible with standard numerical solvers for partial differential equations. The reduced density matrix is obtained from a single deterministic solve at a cost comparable to that of computing one stochastic hierarchy trajectory. Replacing an ensemble of $N_{\mathrm{traj}}$ trajectories therefore reduces the computational cost by approximately a factor of $N_{\mathrm{traj}}$ and eliminates statistical sampling error. Benchmark comparisons show that the proposed method achieves smaller errors at lower computational cost than the tested stochastic hierarchy methods. Applications involving nonquadratic potentials, diffraction and interference, and coupled three-dimensional dynamics demonstrate its ability to handle challenging problems.
AI Summary
Non-Markovian quantum state diffusion (NMQSD) requires handling a functional derivative and averaging stochastic trajectories to recover reduced dynamics. This paper develops a deterministic polynomial chaos Galerkin (DPCG) formulation that replaces trajectory sampling with coupled wavefunction coefficient equations. A rank-$Q$ decomposition of the bath correlation function parameterizes the colored Gaussian noise by $Q$ independent complex Gaussian variables. Expanding the wavefunction in normalized monomials through total degree $P$ and applying Galerkin projection gives $m=\binom{Q+P}{P}$ coefficient wavefunctions. Their outer products sum directly to the approximate reduced density matrix, which is positive semidefinite and trace-preserving before time discretization. At identical hierarchy cutoffs, the coefficient count matches one HOPS or low-rank stochastic hierarchy trajectory. For bounded $H,L$, self-adjoint $H$, and stated regularity and unitary-propagator assumptions, the analysis bounds the state error by $C_1[\sqrt{\tau_Q}+C_2^{P+1}/\sqrt{(P+1)!}]+\delta_{\mathrm{num}}$, where $\tau_Q$ is the discarded covariance eigenvalue sum. Tests compare against PT-TEMPO spin–boson dynamics and exact harmonic-system densities. One harmonic benchmark reports density error $8.87\times10^{-4}$ in $3.07$ seconds, versus $6.94\times10^{-3}$ in $3320.72$ seconds for a 1,000-trajectory low-rank stochastic hierarchy at the same $Q,P,K$. Applications include nonquadratic double wells, slit diffraction and interference, and three-coordinate harmonic dynamics. The stated remaining challenge is combinatorial growth of the polynomial space with bath-mode count and expansion order; adaptive basis selection is proposed as future work. The convergence theorem assumes bounded operators, unlike the unbounded continuous-coordinate operators used in the applications.
Central claim may not hold: The approximately N_traj speedup is justified at fixed hierarchy dimensions and supported by selected benchmarks, but is not established generally at matched accuracy because the deterministic and stochastic hierarchy truncations are different approximations.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.