Taishan Zeng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Taishan Zeng has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Bihui Yu
Chinese Academy of Sciences
1 shared publication - Chao Deng
School of Computer Science and Engineering, Central South University, Changsha Hunan Provincial Key Lab on Bioinformatics, Central South University, Changsha
1 shared publication Chunyuan Lu
Guangdong Pharmaceut Univ, Coll Med Informat Engn, Guangzhou 510006, Peoples R China
1 shared publicationDongsheng Cheng
Shenzhen Inst Informat Technol, Sch Software Engn, Shenzhen 518172, Peoples R China
1 shared publication- Jian Chen
Department of Mathematics, Foshan University, Foshan 528000, China
1 shared publication - Jianwei Cao
South China Normal University, Guangzhou, China
1 shared publication
Latest Publications
- 2 November 2026
- Published by Springer Nature in Multimedia Systems
- Vol. 32 (11) , 682
The explain-on-blackboard paradigm for interleaved visual-text reasoning
- 2 November 2026
- Published by Springer Nature in Multimedia Systems
- Vol. 32 (11) , 682
The explain-on-blackboard paradigm for interleaved visual-text reasoning
- 12 December 2024
- Published by Elsevier in Journal of Computational Physics
- Vol. 523
Sparse-regularized high-frequency enhanced neural network for solving high-frequency problems
- 12 December 2024
- Published by Elsevier in Journal of Computational Physics
- Vol. 523
Sparse-regularized high-frequency enhanced neural network for solving high-frequency problems
- 12 April 2024
- Published by Elsevier in Applied Mathematics Letters
- Vol. 154
A phase velocity preserving fourth-order finite difference scheme for the Helmholtz equation with variable wavenumber
- 12 April 2024
- Published by Elsevier in Applied Mathematics Letters
- Vol. 154
A phase velocity preserving fourth-order finite difference scheme for the Helmholtz equation with variable wavenumber
- Open Access
Multi-Grade Deep Learning for Partial Differential Equations with Applications to the Burgers Equation
We develop in this paper a multi-grade deep learning method for solving nonlinear partial differential equations (PDEs). Deep neural networks (DNNs) have received super performance in solving PDEs in addition to their outstanding success in areas such as natural language processing, computer vision, and robotics. However, training a very deep network is often a challenging task. As the number of layers of a DNN increases, solving a large-scale non-convex optimization problem that results in the DNN solution of PDEs becomes more and more difficult, which may lead to a decrease rather than an increase in predictive accuracy. To overcome this challenge, we propose a two-stage multi-grade deep learning (TS-MGDL) method that breaks down the task of learning a DNN into several neural networks stacked on top of each other in a staircase-like manner. This approach allows us to mitigate the complexity of solving the non-convex optimization problem with large number of parameters and learn residual components left over from previous grades efficiently. We prove that each grade/stage of the proposed TS-MGDL method can reduce the value of the loss function and further validate this fact through numerical experiments. Although the proposed method is applicable to general PDEs, implementation in this paper focuses only on the 1D, 2D, and 3D viscous Burgers equations. Experimental results show that the proposed two-stage multi-grade deep learning method enables efficient learning of solutions of the equations and outperforms existing single-grade deep learning methods in predictive accuracy. Specifically, the predictive errors of the single-grade deep learning are larger than those of the TS-MGDL method in 26-60, 4-31 and 3-12 times, for the 1D, 2D, and 3D equations, respectively.
Open AccessMulti-Grade Deep Learning for Partial Differential Equations with Applications to the Burgers Equation
We develop in this paper a multi-grade deep learning method for solving nonlinear partial differential equations (PDEs). Deep neural networks (DNNs) have received super performance in solving PDEs in addition to their outstanding success in areas such as natural language processing, computer vision, and robotics. However, training a very deep network is often a challenging task. As the number of layers of a DNN increases, solving a large-scale non-convex optimization problem that results in the DNN solution of PDEs becomes more and more difficult, which may lead to a decrease rather than an increase in predictive accuracy. To overcome this challenge, we propose a two-stage multi-grade deep learning (TS-MGDL) method that breaks down the task of learning a DNN into several neural networks stacked on top of each other in a staircase-like manner. This approach allows us to mitigate the complexity of solving the non-convex optimization problem with large number of parameters and learn residual components left over from previous grades efficiently. We prove that each grade/stage of the proposed TS-MGDL method can reduce the value of the loss function and further validate this fact through numerical experiments. Although the proposed method is applicable to general PDEs, implementation in this paper focuses only on the 1D, 2D, and 3D viscous Burgers equations. Experimental results show that the proposed two-stage multi-grade deep learning method enables efficient learning of solutions of the equations and outperforms existing single-grade deep learning methods in predictive accuracy. Specifically, the predictive errors of the single-grade deep learning are larger than those of the TS-MGDL method in 26-60, 4-31 and 3-12 times, for the 1D, 2D, and 3D equations, respectively.
- 14 January 2023
- Published by Elsevier in Applied Mathematics Letters
- Vol. 140
Numerical analysis of the optimal 9-point finite difference scheme for the Helmholtz equation
- 14 January 2023
- Published by Elsevier in Applied Mathematics Letters
- Vol. 140
Numerical analysis of the optimal 9-point finite difference scheme for the Helmholtz equation
- Open Access
- 14 December 2021
- Published by MDPI AG in Sustainability
- Vol. 13 (24) , 13770
Crude Oil Price Forecast Based on Deep Transfer Learning: Shanghai Crude Oil as an Example
Crude oil is an important fuel resource for all countries. Accurate predictions of oil prices have important economic and social values. However, the price of crude oil is highly nonlinear under the influence of many factors, so it is very difficult to predict accurately. Shanghai crude oil futures were officially listed in March 2018. It is of great significance to accurately predict the price of Shanghai crude oil futures for guiding China’s domestic production practice. Forecasting the price of Shanghai crude oil futures is even more difficult because of the lack of price data due to the short listing time. In order to solve this problem, this paper proposes using Long Short-Term Memory Network (LSTM) based on transfer learning to predict the price of crude oil in Shanghai. The basic idea is to take advantage of the correlation between Brent crude oil and Shanghai crude oil, use Brent crude oil for training in the early stage, and then use Shanghai crude oil to fine-tune the network. The empirical results show that the LSTM model based on transfer learning has strong generalization ability and high prediction accuracy.
- 14 December 2021
- Published by MDPI AG in Sustainability
- Vol. 13 (24) , 13770
Open AccessCrude Oil Price Forecast Based on Deep Transfer Learning: Shanghai Crude Oil as an Example
Crude oil is an important fuel resource for all countries. Accurate predictions of oil prices have important economic and social values. However, the price of crude oil is highly nonlinear under the influence of many factors, so it is very difficult to predict accurately. Shanghai crude oil futures were officially listed in March 2018. It is of great significance to accurately predict the price of Shanghai crude oil futures for guiding China’s domestic production practice. Forecasting the price of Shanghai crude oil futures is even more difficult because of the lack of price data due to the short listing time. In order to solve this problem, this paper proposes using Long Short-Term Memory Network (LSTM) based on transfer learning to predict the price of crude oil in Shanghai. The basic idea is to take advantage of the correlation between Brent crude oil and Shanghai crude oil, use Brent crude oil for training in the early stage, and then use Shanghai crude oil to fine-tune the network. The empirical results show that the LSTM model based on transfer learning has strong generalization ability and high prediction accuracy.
- Open Access
- 2 March 2021
- Published by Wiley in Journal of Function Spaces
- Vol. 2021, 1-11
L 1 -Multiscale Galerkin’s Scheme with Multilevel Augmentation Algorithm for Solving Time Fractional Burgers’ Equation
In this paper, we consider the initial boundary value problem of the time fractional Burgers equation. A fully discrete scheme is proposed for the time fractional nonlinear Burgers equation with time discretized by -type formula and space discretized by the multiscale Galerkin method. The optimal convergence orders reach in the norm and in the norm, respectively, in which is the time step size, is the space step size, and is the order of piecewise polynomial space. Then, a fast multilevel augmentation method (MAM) is developed for solving the nonlinear algebraic equations resulting from the fully discrete scheme at each time step. We show that the MAM preserves the optimal convergence orders, and the computational cost is greatly reduced. Numerical experiments are presented to verify the theoretical analysis, and comparisons between MAM and Newton’s method show the efficiency of our algorithm.
- 2 March 2021
- Published by Wiley in Journal of Function Spaces
- Vol. 2021, 1-11
Open AccessL 1 -Multiscale Galerkin’s Scheme with Multilevel Augmentation Algorithm for Solving Time Fractional Burgers’ Equation
In this paper, we consider the initial boundary value problem of the time fractional Burgers equation. A fully discrete scheme is proposed for the time fractional nonlinear Burgers equation with time discretized by -type formula and space discretized by the multiscale Galerkin method. The optimal convergence orders reach in the norm and in the norm, respectively, in which is the time step size, is the space step size, and is the order of piecewise polynomial space. Then, a fast multilevel augmentation method (MAM) is developed for solving the nonlinear algebraic equations resulting from the fully discrete scheme at each time step. We show that the MAM preserves the optimal convergence orders, and the computational cost is greatly reduced. Numerical experiments are presented to verify the theoretical analysis, and comparisons between MAM and Newton’s method show the efficiency of our algorithm.