Yingqi Tian
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Yingqi Tian has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Haibo Ma
Shandong University
3 shared publications Zhaoxuan Xie
Nanjing University
3 shared publications- Zhong Jin
Key Laboratory of Functional Molecular Solids College of Chemistry and Molecular Sciences Ministry of Education Anhui Normal University Wuhu Anhui China
3 shared publications - Bingbing Suo
Shaanxi Key Laboratory for Theoretical Physics Frontiers, Institute of Modern Physics
2 shared publications - Honghui Shang
University of Science and Technology of China
2 shared publications - Yingjin Ma
Chinese Academy of Sciences
2 shared publications
Latest Publications
- 15 August 2026
- Published by Springer Nature
- p. 481-494
TAIN: Scalable Tensor-Aware Acceleration for Quantum Chemistry on Heterogeneous Architectures
- 15 August 2026
- Published by Springer Nature
- p. 481-494
TAIN: Scalable Tensor-Aware Acceleration for Quantum Chemistry on Heterogeneous Architectures
- 2 March 2025
- Published by American Chemical Society (ACS) in Journal of Chemical Theory and Computation
- Vol. 21 (7) , 3426-3439
Bridging the Gap between Transformer-Based Neural Networks and Tensor Networks for Quantum Chemistry
The neural network quantum state (NNQS) method has demonstrated promising results in ab initio quantum chemistry, achieving remarkable accuracy in molecular systems. However, efficient calculation of systems with large active spaces remains challenging. This study introduces a novel approach that bridges tensor network states with the transformer-based NNQS-Transformer (QiankunNet) to enhance accuracy and convergence for systems with relatively large active spaces. By transforming tensor network states into active space configuration interaction type wave functions, QiankunNet achieves accuracy surpassing both the pretraining density matrix renormalization group (DMRG) results and traditional coupled cluster methods, particularly in strongly correlated regimes. We investigate two configuration transformation methods: the sweep-based direct conversion (Conv.) method and the entanglement-driven genetic algorithm (EDGA) method, with Conv. showing superior efficiency. The effectiveness of this approach is validated on H2O with a large active space (10e, 24o) in the cc-pVDZ basis set, demonstrating an efficient routine between DMRG and QiankunNet and also offering a promising direction for advancing quantum state representation in complex molecular systems.
- 2 March 2025
- Published by American Chemical Society (ACS) in Journal of Chemical Theory and Computation
- Vol. 21 (7) , 3426-3439
Bridging the Gap between Transformer-Based Neural Networks and Tensor Networks for Quantum Chemistry
The neural network quantum state (NNQS) method has demonstrated promising results in ab initio quantum chemistry, achieving remarkable accuracy in molecular systems. However, efficient calculation of systems with large active spaces remains challenging. This study introduces a novel approach that bridges tensor network states with the transformer-based NNQS-Transformer (QiankunNet) to enhance accuracy and convergence for systems with relatively large active spaces. By transforming tensor network states into active space configuration interaction type wave functions, QiankunNet achieves accuracy surpassing both the pretraining density matrix renormalization group (DMRG) results and traditional coupled cluster methods, particularly in strongly correlated regimes. We investigate two configuration transformation methods: the sweep-based direct conversion (Conv.) method and the entanglement-driven genetic algorithm (EDGA) method, with Conv. showing superior efficiency. The effectiveness of this approach is validated on H2O with a large active space (10e, 24o) in the cc-pVDZ basis set, demonstrating an efficient routine between DMRG and QiankunNet and also offering a promising direction for advancing quantum state representation in complex molecular systems.
- Open Access
- 29 January 2024
- Published by MDPI AG in Mathematics
- Vol. 12 (3) , 433
Solving the Electronic Schrödinger Equation by Pairing Tensor-Network State with Neural Network Quantum State
Neural network methods have shown promise for solving complex quantum many-body systems. In this study, we develop a novel approach through incorporating the density-matrix renormalization group (DMRG) method with the neural network quantum state method. The results demonstrate that, when tensor-network pre-training is introduced into the neural network, a high efficiency can be achieved for quantum many-body systems with strong correlations.
- 29 January 2024
- Published by MDPI AG in Mathematics
- Vol. 12 (3) , 433
Open AccessSolving the Electronic Schrödinger Equation by Pairing Tensor-Network State with Neural Network Quantum State
Neural network methods have shown promise for solving complex quantum many-body systems. In this study, we develop a novel approach through incorporating the density-matrix renormalization group (DMRG) method with the neural network quantum state method. The results demonstrate that, when tensor-network pre-training is introduced into the neural network, a high efficiency can be achieved for quantum many-body systems with strong correlations.
- 9 October 2023
- Published by American Physical Society (APS) in Physical Review Letters
- Vol. 131 (15) , 158001
Mechanistic Insights into Water Autoionization through Metadynamics Simulation Enhanced by Machine Learning
Characterizing the free energy landscape of water ionization has been a great challenge due to the limitations from expensive ab initio calculations and strong rare-event features. Lacking equilibrium sampling of the ionization pathway will cause ambiguities in the mechanistic study. Here, we obtain convergent free energy surfaces through nanosecond timescale metadynamics simulations with classical nuclei enhanced by atomic neural network potentials, which yields good reproduction of the equilibrium constant () and ionization rate constant (). The character of transition state unveils the triple-proton transfer occurs through a concerted but asynchronous mechanism. Conditional ensemble average analyses establish the dual-presolvation mechanism, where a pair of hypercoordinated and undercoordinated waters bridged by one cooperatively constitutes the initiation environment for autoionization, and contributes extremely to the local electric field fluctuation to promote water dissociation.- 9 October 2023
- Published by American Physical Society (APS) in Physical Review Letters
- Vol. 131 (15) , 158001
Mechanistic Insights into Water Autoionization through Metadynamics Simulation Enhanced by Machine Learning
Characterizing the free energy landscape of water ionization has been a great challenge due to the limitations from expensive ab initio calculations and strong rare-event features. Lacking equilibrium sampling of the ionization pathway will cause ambiguities in the mechanistic study. Here, we obtain convergent free energy surfaces through nanosecond timescale metadynamics simulations with classical nuclei enhanced by atomic neural network potentials, which yields good reproduction of the equilibrium constant () and ionization rate constant (). The character of transition state unveils the triple-proton transfer occurs through a concerted but asynchronous mechanism. Conditional ensemble average analyses establish the dual-presolvation mechanism, where a pair of hypercoordinated and undercoordinated waters bridged by one cooperatively constitutes the initiation environment for autoionization, and contributes extremely to the local electric field fluctuation to promote water dissociation.- 1 June 2023
- Published by Bentham Science Publishers Ltd. in Current Chinese Science
- Vol. 3 (3) , 178-186
High-Performance Computing for Density Matrix Renormalization Group
In the last decades, many algorithms have been developed to use high-performance computing (HPC) techniques to accelerate the density matrix renormalization group (DMRG) method, an effective method for solving large active space strong correlation problems. In this article, the previous DMRG parallelization algorithms at different levels of the parallelism are introduced. The heterogeneous computing acceleration methods and the mixed-precision implementation are also presented and discussed. This mini-review concludes with some summary and prospects for future works.
- 1 June 2023
- Published by Bentham Science Publishers Ltd. in Current Chinese Science
- Vol. 3 (3) , 178-186
High-Performance Computing for Density Matrix Renormalization Group
In the last decades, many algorithms have been developed to use high-performance computing (HPC) techniques to accelerate the density matrix renormalization group (DMRG) method, an effective method for solving large active space strong correlation problems. In this article, the previous DMRG parallelization algorithms at different levels of the parallelism are introduced. The heterogeneous computing acceleration methods and the mixed-precision implementation are also presented and discussed. This mini-review concludes with some summary and prospects for future works.
- Open Access
- 29 November 2022
- Published by American Chemical Society (ACS) in ChemRxiv
Kylin 1.0: An Ab-Initio Density Matrix Renormalization Group Quantum Chemistry Program
The accurate evaluation of electron correlations is highly necessary for the proper descriptions of the electronic structures in strongly correlated molecules, ranging from bond-dissociating molecules, polyradicals, to large conjugated molecules and transition metal complexes. For this purpose, in this paper, a new ab-initio quantum chemistry program Kylin 1.0 for electron correlation calculations at various quantum many-body levels, including configuration interaction (CI), perturbation theory (PT), and density matrix renormalization group (DMRG), is presented. In addtion, fundamental quantum chemical methods such as Hartree-Fock self-consistent field (HF-SCF) and the complete active space SCF (CASSCF) are aslo implemented. The Kylin 1.0 program possesses these features: (1) efficient DMRG implementation based on the matrix product operator (MPO) formulation for describing static electron correlation within a large active space composed of more than 100 orbitals, supporting both $\rm U(1)_{n} \times U(1)_{S_z}$ and $\rm U(1)_{n} \times SU(2)_{S}$ symmetries; (2) efficient second-order DMRG-self-consistent field (SCF) implementation; (3) externally-contracted multi-reference CI (MRCI) and Epstein-Nesbet PT with DMRG reference wave functions for including the remaining dynamic electron correlation outside the large active spaces. In this paper, we introduce the capabilities and numerical benchmark examples of the Kylin 1.0 program.- 29 November 2022
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessKylin 1.0: An Ab-Initio Density Matrix Renormalization Group Quantum Chemistry Program
The accurate evaluation of electron correlations is highly necessary for the proper descriptions of the electronic structures in strongly correlated molecules, ranging from bond-dissociating molecules, polyradicals, to large conjugated molecules and transition metal complexes. For this purpose, in this paper, a new ab-initio quantum chemistry program Kylin 1.0 for electron correlation calculations at various quantum many-body levels, including configuration interaction (CI), perturbation theory (PT), and density matrix renormalization group (DMRG), is presented. In addtion, fundamental quantum chemical methods such as Hartree-Fock self-consistent field (HF-SCF) and the complete active space SCF (CASSCF) are aslo implemented. The Kylin 1.0 program possesses these features: (1) efficient DMRG implementation based on the matrix product operator (MPO) formulation for describing static electron correlation within a large active space composed of more than 100 orbitals, supporting both $\rm U(1)_{n} \times U(1)_{S_z}$ and $\rm U(1)_{n} \times SU(2)_{S}$ symmetries; (2) efficient second-order DMRG-self-consistent field (SCF) implementation; (3) externally-contracted multi-reference CI (MRCI) and Epstein-Nesbet PT with DMRG reference wave functions for including the remaining dynamic electron correlation outside the large active spaces. In this paper, we introduce the capabilities and numerical benchmark examples of the Kylin 1.0 program. - 2 November 2022
- Published by American Chemical Society (ACS) in Journal of Chemical Theory and Computation
- Vol. 18 (12) , 7260-7271
Mixed-Precision Implementation of the Density Matrix Renormalization Group
The mixed-precision optimization is an effective emerging technique for quantum chemistry methods to obtain better computational performance and maintain the chemical accuracy. Here, we developed a two-level mixed-precision implementation for the density matrix renormalization group (DMRG) method. This implementation is based on the idea that the DMRG process is an iterative process. Therefore, the first several iteration steps can be executed in single precision. A feasible single-precision DMRG may generate moderate accuracy, and when a few double-precision cleanup sweeps are added, the double-precision accuracy will be recovered. In the double precision sweeps, we developed a mixed-precision diagonalization method that can run the most time-consuming step in single precision and maintain the double-precision accuracy. By combining these two mixed-precision schemes, we implemented our mixed-precision DMRG method. The benchmark result shows that our mixed-precision implementation achieved a good performance. A speed-up of up to 2.31 is achieved, and the accuracy is preserved within 0.01 kcal/mol.
- 2 November 2022
- Published by American Chemical Society (ACS) in Journal of Chemical Theory and Computation
- Vol. 18 (12) , 7260-7271
Mixed-Precision Implementation of the Density Matrix Renormalization Group
The mixed-precision optimization is an effective emerging technique for quantum chemistry methods to obtain better computational performance and maintain the chemical accuracy. Here, we developed a two-level mixed-precision implementation for the density matrix renormalization group (DMRG) method. This implementation is based on the idea that the DMRG process is an iterative process. Therefore, the first several iteration steps can be executed in single precision. A feasible single-precision DMRG may generate moderate accuracy, and when a few double-precision cleanup sweeps are added, the double-precision accuracy will be recovered. In the double precision sweeps, we developed a mixed-precision diagonalization method that can run the most time-consuming step in single precision and maintain the double-precision accuracy. By combining these two mixed-precision schemes, we implemented our mixed-precision DMRG method. The benchmark result shows that our mixed-precision implementation achieved a good performance. A speed-up of up to 2.31 is achieved, and the accuracy is preserved within 0.01 kcal/mol.
- Open Access
- 15 June 2022
- Published by American Chemical Society (ACS) in ChemRxiv
A Mixed-Precision Implementation of the Density Matrix Renormalization Group
Using the mixed precision strategy to optimize quantum chemistry codes has been proved promising in saving computational cost and maintaining chemical accuracy. Here, an efficient mixed-precision density matrix renormalization group (DMRG) scheme, containing a two-level mixed-precision hierarchy, is developed and demonstrated. At the coarse-grained level, based on the discovery that the single-precision orthogonalization may cause the DMRG generate a totally wrong answer, a feasible single-precision-sweep DMRG method with double-precision orthogonalization process is implemented. At the fine-grained level, a mixed-precision diagonalization algorithm is developed. This algorithm runs specific operations in the single-precision while preserving double-precision accuracy. Combining these two method, a hybrid mixed-precision scheme is presented. By applying this scheme, the DMRG single-point energy calculations are accelerated up to 131%. Mixed-precision DMRG yielded energies are accurate and deviate less than 0.01 kcal/mol compared with standard DMRG calculations.
- 15 June 2022
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessA Mixed-Precision Implementation of the Density Matrix Renormalization Group
Using the mixed precision strategy to optimize quantum chemistry codes has been proved promising in saving computational cost and maintaining chemical accuracy. Here, an efficient mixed-precision density matrix renormalization group (DMRG) scheme, containing a two-level mixed-precision hierarchy, is developed and demonstrated. At the coarse-grained level, based on the discovery that the single-precision orthogonalization may cause the DMRG generate a totally wrong answer, a feasible single-precision-sweep DMRG method with double-precision orthogonalization process is implemented. At the fine-grained level, a mixed-precision diagonalization algorithm is developed. This algorithm runs specific operations in the single-precision while preserving double-precision accuracy. Combining these two method, a hybrid mixed-precision scheme is presented. By applying this scheme, the DMRG single-point energy calculations are accelerated up to 131%. Mixed-precision DMRG yielded energies are accurate and deviate less than 0.01 kcal/mol compared with standard DMRG calculations.
- 20 July 2021
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 155 (3) , 034112
Optimizing two-electron repulsion integral calculations with McMurchie–Davidson method on graphic processing unit
In this article, several optimization methods of two-electron repulsion integral calculations on a graphic processing unit (GPU) are presented. These methods are based on the investigations of the method presented by McMurchie and Davidson (MD). A new Boys function evaluation method for the GPU calculation is introduced. The series summation, the error function, and the finite sum formula method are combined; thus, good performance on the GPU can be achieved. By taking some theoretical study of the McMurchie–Davidson recurrence relations, three major optimization approaches are derived from the deduction of the general term formula for the Hermite expansion coefficient. The three approaches contain a new form of the Hermite expansion coefficients with corresponding recurrence relations, which is more efficient for one-electron integrals and [ss|∗∗] or [∗∗|ss] type two-electron integrals. In addition, a simple yet efficient new recurrence formula for the coefficient evaluation is derived, which is more efficient both in float operations and memory operations than its original one. In average, the new recurrence relation can save 26% float operations and 37% memory operations. Finally, a common sub-expression elimination (CSE) method is implemented. This CSE method is directly generated from some equalities we discovered from the general term formula other than by computer algebra system software. This optimized method achieved up to 3.09 speedups compared to the original MD method on the GPU and up to 92.75 speedups compared to the GAMESS calculation on the central processing unit.
- 20 July 2021
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 155 (3) , 034112
Optimizing two-electron repulsion integral calculations with McMurchie–Davidson method on graphic processing unit
In this article, several optimization methods of two-electron repulsion integral calculations on a graphic processing unit (GPU) are presented. These methods are based on the investigations of the method presented by McMurchie and Davidson (MD). A new Boys function evaluation method for the GPU calculation is introduced. The series summation, the error function, and the finite sum formula method are combined; thus, good performance on the GPU can be achieved. By taking some theoretical study of the McMurchie–Davidson recurrence relations, three major optimization approaches are derived from the deduction of the general term formula for the Hermite expansion coefficient. The three approaches contain a new form of the Hermite expansion coefficients with corresponding recurrence relations, which is more efficient for one-electron integrals and [ss|∗∗] or [∗∗|ss] type two-electron integrals. In addition, a simple yet efficient new recurrence formula for the coefficient evaluation is derived, which is more efficient both in float operations and memory operations than its original one. In average, the new recurrence relation can save 26% float operations and 37% memory operations. Finally, a common sub-expression elimination (CSE) method is implemented. This CSE method is directly generated from some equalities we discovered from the general term formula other than by computer algebra system software. This optimized method achieved up to 3.09 speedups compared to the original MD method on the GPU and up to 92.75 speedups compared to the GAMESS calculation on the central processing unit.
- Open Access
- 1 January 2021
- Published by Shanghai Institute of Organic Chemistry in Acta Chimica Sinica
- Vol. 79 (5) , 653-658
Hartree-Fock and Density Functional Calculations on Graphics Processing Unit
- 1 January 2021
- Published by Shanghai Institute of Organic Chemistry in Acta Chimica Sinica
- Vol. 79 (5) , 653-658
Open AccessHartree-Fock and Density Functional Calculations on Graphics Processing Unit