One-Shot Trajectory Learning of Open Quantum Systems Dynamics

Abstract

Nonadiabatic quantum dynamics is important for understanding light-harvesting processes, but its propagation with traditional methods can be rather expensive. Here we present a one-shot trajectory learning approach that allows us to directly make an ultrafast prediction of the entire trajectory of the reduced density matrix for a new set of such simulation parameters as temperature and reorganization energy. The whole 10-ps-long propagation takes 70 ms as we demonstrate on the comparatively large quantum system, the Fenna–Matthews–Olsen (FMO) complex. Our approach also significantly reduces time and memory requirements for training.

A comparative study of different machine learning methods for dissipative quantum dynamics

“…Recently, CNNs is used to study the excitation energy transfer in Fenna-Matthews-Olson light-harvesting complex. 34,35 Wu et al 122 used a hybrid CNN/LSTM network to predict long-time semiclassical and mixed quantum-classical dynamics of the spin-boson model. Lin et al 123,130 trained a multi-layer LSTM model to simulate the long-time dynamics of spin-boson model and used bootstrap method to estimate the confidence interval.…”

“…Machine learning (ML) offers an alternative route to accurate, yet greatly accelerated quantum dynamics calculations with minimum input information required. [32][33][34][35] In ML, predicting the future time-evolution of quantum mechanical observables based on past values can be formulated as time series forecasting problem. A time series is a set of data points recorded in consecutive time intervals.…”

“…Recently, many ML models have been applied to simulate the dynamics of quantum systems. [32][33][34][35][117][118][119][120][121][122][123][124][125][126] We note that ML can also be applied to quantum dynamics in a different context-namely as surrogate models for quantum chemical properties such as potential energies and forces in different electronic states as well as cou-plings between states eliminating the need for expensive (excited-state) electronic structure calculations. [127][128][129] Here we apply ML to propagate a quantum system assuming potential energies are readily available.…”

A comparative study of different machine learning methods for dissipative quantum dynamics

“…Machine learning (ML) offers an alternative route to accurate, yet greatly accelerated quantum dynamics calculations with minimum input information required. [32][33][34][35] In ML, predicting the future time-evolution of quantum mechanical observables based on past values can be formulated as time series forecasting problem. A time series is a set of data points recorded in consecutive time intervals.…”

“…116 Recently, many ML models have been applied to simulate the dynamics of quantum systems. [32][33][34][35][117][118][119][120][121][122][123][124][125][126] We note that ML can also be applied to quantum dynamics in a different context-namely as surrogate models for quantum chemical properties such as potential energies and forces in different electronic states as well as couplings between the states eliminating the need for expensive (excited-state) electronic structure calculations. [127][128][129] Here we apply ML to propagate a quantum system assuming potential energies are readily available.…”