Speeding up quantum dissipative dynamics of open systems with kernel methods

Abstract

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.

One-shot trajectory learning of open quantum systems dynamics

“…The mean absolute errors averaged over 1380 hold-out test trajectories for initial excitation on site-1 and site-6 are 4.19 • 10 −4 and 4.59 • 10 −4 , respectively, i.e., rather small. Prediction of the 2.5 ps long trajectory takes less than 50 millisecond; for comparison, the same propagation with our previous AI-QD approach [14] takes ca. 2 min on the same machine (single core of Intel(R) Core(TM) i7-10700 CPUs @ 2.90 GHz).…”

“…Recently, ML has been successfully applied to accelerate dynamics propagation of open quantum systems. [10][11][12][13][14][15][16][17][18][19] ML-based approaches adopted in the literature so far, can be divided into two categories: recursive [13][14][15][16] and non-recursive [17,18] approaches. ML-based recursive approaches are quite successful but there are some downsides of them; first, iterative propagation is inherently slow and may lead to error accumulation and second, they need a short time-trajectory generated with traditional quantum dynamics methods.…”

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

“…Using the vanishing gradient scheme to find different t M for each trajectory allows us to sample more data from the training trajectories, which are hard-to-learn, while avoiding redundant sampling from trajectories, which are easy-to-learn. This also removes arbitrariness in choosing fixed t M parameter as was done in previous studies using the recursive AI-QD scheme 44 , 46 .…”

“…Alleviating the computational cost of QD became a target of a series of studies applying artificial intelligence (AI) 40 – 46 , inspired by advances in application of AI employing machine learning (ML) algorithms in computational chemistry and chemical physics 47 , 48 . AI was also applied to investigate EET in a dimer system 44 and the FMO complex 40 .…”

“…AI was also applied to investigate EET in a dimer system 44 and the FMO complex 40 . Saving of computational cost by AI in above studies is impressive, however, one of the studies 40 only focused on predicting energy transfer times and transfer efficiencies rather than temporal and spatial evolution, while other related studies 44 – 46 adopted basically the same recursive nature of QD trajectory propagation.…”