The future forecasting ability of machine learning (ML) makes ML a promising tool for predicting long-time quantum dissipative dynamics of open systems. In this Article, we employ nonparametric machine learning algorithm (kernel ridge regression as a representative of the kernel methods) to study the quantum dissipative dynamics of the widely-used spin-boson model. Our ML model takes short-time dynamics as an input and is used for fast propagation of the long-time dynamics, greatly reducing the computational effort in comparison with the traditional approaches. Presented results show that the ML model performs well in both symmetric and asymmetric spin-boson models. Our approach is not limited to spin-boson model and can be extended to complex systems.

Speeding up quantum dissipative dynamics of open systems with kernel methods

The core objective of machine-assisted scientific discovery is to learn physical laws from experimental data without prior knowledge of the systems in question. In the area of quantum physics, making progress towards these goals is significantly more challenging due to the curse of dimensionality as well as the counterintuitive nature of quantum mechanics. Here we present the QNODE, a latent neural ordinary differential equation (ODE) trained on expectation values of closed and open-quantum-systems dynamics. It can learn to generate such measurement data and extrapolate outside of its training region that satisfies the von Neumann and time-local Lindblad master equations for closed and open quantum systems, respectively, in an unsupervised means. Furthermore, the QNODE rediscovers quantum-mechanical laws such as the Heisenberg's uncertainty principle in a data-driven way, without any constraint or guidance. Additionally, we show that trajectories that are generated from the QNODE that are close in its latent space have similar quantum dynamics while preserving the physics of the training system. 

/pdf/learning-quantum-dynamics-with-latent-neural-ordinary-3nrkzqtg.pdf

Learning quantum dynamics with latent neural ordinary differential equations 

Exploring excitation energy transfer (EET) in light-harvesting complexes (LHCs) is essential for understanding the natural processes and design of highly-efficient photovoltaic devices. LHCs are open systems, where quantum effects may play a crucial role for almost perfect utilization of solar energy. Simulation of energy transfer with inclusion of quantum effects can be done within the framework of dissipative quantum dynamics (QD), which are computationally expensive. Thus, artificial intelligence (AI) offers itself as a tool for reducing the computational cost. Here we suggest AI-QD approach using AI to directly predict QD as a function of time and other parameters such as temperature, reorganization energy, etc., completely circumventing the need of recursive step-wise dynamics propagation in contrast to the traditional QD and alternative, recursive AI-based QD approaches. Our trajectory-learning AI-QD approach is able to predict the correct asymptotic behavior of QD at infinite time. We demonstrate AI-QD on seven-sites Fenna-Matthews-Olson (FMO) complex. 

/pdf/predicting-the-future-of-excitation-energy-transfer-in-light-2in6g39q.pdf

Predicting the future of excitation energy transfer in light-harvesting complex with artificial intelligence-based quantum dynamics

We present a series of new implementations that we recently introduced in COBRAMM, the open‐source academic software developed in our group. The goal of these implementations is to offer an automatized workflow and interface to simulate time‐resolved transient absorption (TA) spectra of medium‐to‐big chromophore embedded in a complex environment. Therefore, the excited states absorption and the stimulated emission are simulated along nonadiabatic dynamics performed with trajectory surface hopping. The possibility of treating systems from medium to big size is given by the use of time‐dependent density functional theory (TD‐DFT) and the presence of the environment is taken into account employing a hybrid quantum mechanics/molecular mechanics (QM/MM) scheme. The full implementation includes a series of auxiliary scripts to properly setup the QM/MM system, the calculation of the wavefunction overlap along the dynamics for the propagation, the evaluation of the transition dipole moment at linear response TD‐DFT level, and scripts to setup, run and analyze the TA from an ensemble of trajectories. Altogether, we believe that our implementation will open the door to the easily simulate the time‐resolved TA of systems so far computationally inaccessible.

https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/jcc.26966

Automatized protocol and interface to simulate QM/MM time‐resolved transient absorption at TD‐DFT level with COBRAMM

Femtosecond nonlinear spectroscopy is the main tool for the time-resolved detection of photophysical and photochemical processes. Since most systems of chemical interest are rather complex, theoretical support is indispensable for the extraction of the intrinsic system dynamics from the detected spectroscopic responses. There exist two alternative theoretical formalisms for the calculation of spectroscopic signals, the nonlinear response-function (NRF) approach and the spectroscopic equation-of-motion (EOM) approach. In the NRF formalism, the system-field interaction is assumed to be sufficiently weak and is treated in lowest-order perturbation theory for each laser pulse interacting with the sample. The conceptual alternative to the NRF method is the extraction of the spectroscopic signals from the solutions of quantum mechanical, semiclassical, or quasiclassical EOMs which govern the time evolution of the material system interacting with the radiation field of the laser pulses. The NRF formalism and its applications to a broad range of material systems and spectroscopic signals have been comprehensively reviewed in the literature. This article provides a detailed review of the suite of EOM methods, including applications to 4-wave-mixing and N-wave-mixing signals detected with weak or strong fields. Under certain circumstances, the spectroscopic EOM methods may be more efficient than the NRF method for the computation of various nonlinear spectroscopic signals.

Equation-of-Motion Methods for the Calculation of Femtosecond Time-Resolved 4-Wave-Mixing and N-Wave-Mixing Signals.

The recurrent neural network with the long short-term memory cell (LSTM-NN) is employed to simulate the long-time dynamics of open quantum systems. The bootstrap method is applied in the LSTM-NN construction and prediction, which provides a Monte Carlo estimation of a forecasting confidence interval. Within this approach, a large number of LSTM-NNs are constructed by resampling time-series sequences that were obtained from the early stage quantum evolution given by numerically exact multilayer multiconfigurational time-dependent Hartree method. The built LSTM-NN ensemble is used for the reliable propagation of the long-time quantum dynamics, and the simulated result is highly consistent with the exact evolution. The forecasting uncertainty that partially reflects the reliability of the LSTM-NN prediction is also given. This demonstrates the bootstrap-based LSTM-NN approach is a practical and powerful tool to propagate the long-time quantum dynamics of open systems with high accuracy and low computational cost.

Simulation of Open Quantum Dynamics with Bootstrap-Based Long Short-Term Memory Recurrent Neural Network.

The ultrafast nonadiabatic internal conversion in azomethane is explored by the on-the-fly trajectory surface-hopping simulations of photoinduced dynamics and femtosecond transient absorption (TA) pump-probe (PP) spectra at three electronic-structure theory levels, OM2/MRCI, SA-CASSCF, and XMS-CASPT2. All these dynamics simulations predict ultrafast internal conversion. On the one hand, the OM2/MRCI and SA-CASSCF methods yield similar excited-state dynamics, while the XMS-CASPT2 method predicts a much slower population decay. On the other hand, the TA PP signals simulated at the SA-CASSCF and XMS-CASPT2 levels show the similar spectral features, particularly for the similar stimulated emission contributions, while the OM2/MRCI signals are quite different. This demonstrates that the nonadiabatic population dynamics and time-resolved stimulated emission signals may reflect different aspects of photoinduced processes. The combination of the dynamical and spectral simulations definitely provides more accurate and detailed information which sheds light on the microscopic mechanisms of photophysical and photochemical processes.

Ultrafast Internal Conversion Dynamics Through the on-the-fly Simulation of Transient Absorption Pump-Probe Spectra with Different Electronic Structure Methods

The machine learning approaches are applied in the dynamical simulation of open quantum systems. The long short-term memory recurrent neural network (LSTM-RNN) models are used to simulate the long-time quantum dynamics, which are built based on the key information of the short-time evolution. We employ various hyperparameter optimization methods, including simulated annealing, Bayesian optimization with tree-structured parzen estimator, and random search, to achieve the automatic construction and adjustment of the LSTM-RNN models. The implementation details of three hyperparameter optimization methods are examined, and among them, the simulated annealing approach is strongly recommended due to its excellent performance. The uncertainties of the machine learning models are comprehensively analyzed by the combination of bootstrap sampling and Monte Carlo dropout approaches, which give the prediction confidence of the LSTM-RNN models in the simulation of the open quantum dynamics. This work builds an effective machine learning approach to simulate the dynamics evolution of open quantum systems. In addition, the current study provides an efficient protocol to build optimal neural networks and estimate the trustiness of the machine learning models.

/pdf/automatic-evolution-of-machine-learning-based-quantum-31ii8dcl.pdf

Automatic Evolution of Machine-Learning-Based Quantum Dynamics with Uncertainty Analysis.

Prediction of the excited-state reaction channels in photo-induced processes of nitrofurantoin using first-principle calculations and dynamics simulations.

The supervised machine learning (ML) approach is applied to realize the trajectory-based nonadiabatic dynamics within the framework of the symmetrical quasi-classical dynamics method based on the Meyer-Miller mapping Hamiltonian (MM-SQC). After the construction of the long short-term memory recurrent neural network (LSTM-RNN) model, it is used to perform the entire trajectory evolutions from initial sampling conditions. The proposed idea is proven to be reliable and accurate in the simulations of the dynamics of several site-exciton electron-phonon coupling models, which cover two-site and three-site systems with biased and unbiased energy levels, as well as include a few or many phonon modes. The LSTM-RNN approach also shows the powerful ability to obtain the accurate and stable results for the long-time evolutions. It indicates that the LSTM-RNN model perfectly captures of dynamical correction information in the trajectory evolution in the MM-SQC dynamics. Our work provides the possibility to employ the ML methods in the simulation of the trajectory-based nonadiabatic dynamic of complex systems with a large number of degrees of freedoms.

Kunni Lin

Papers

Simulation of Open Quantum Dynamics with Bootstrap-Based Long Short-Term Memory Recurrent Neural Network.

Ultrafast Internal Conversion Dynamics Through the on-the-fly Simulation of Transient Absorption Pump-Probe Spectra with Different Electronic Structure Methods

Automatic Evolution of Machine-Learning-Based Quantum Dynamics with Uncertainty Analysis.

Prediction of the excited-state reaction channels in photo-induced processes of nitrofurantoin using first-principle calculations and dynamics simulations.

Realization of the Trajectory Propagation in the MM-SQC Dynamics by Using Machine Learning