Jing-Bin Liang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jing-Bin Liang has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Jianglong Shen
School of Mathematics and Physics, Yibin University, College Street, Yibin 644000, China
4 shared publications - Bao-Ying Du
School of Mathematics and Physics, Yibin University, No. 5, Section 3, Daxue Road, Yibin 644000, China
1 shared publication Gao Ming Deng
1 shared publicationJiang-Long Shen
Yibin University
1 shared publication- Jing-Wen Huang
Yibin University, China
1 shared publication - Jing-Wen Huang
School of Mathematics and Physics, YiBin University, Yibin 644000, China
1 shared publication
Latest Publications
- 2 October 2026
- Published by Elsevier in Ocean Engineering
- Vol. 368
Investigation of multiple soliton solutions for high-dimensional models using automatic logarithmic bilinear transformation
- 2 October 2026
- Published by Elsevier in Ocean Engineering
- Vol. 368
Investigation of multiple soliton solutions for high-dimensional models using automatic logarithmic bilinear transformation
- Open Access
- 12 February 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (4) , 648
A New Neural Network Framework Integrating Symbolic Computation to Solve the (2+1)-Dimensional Boussinesq Equation
The (2+1)-dimensional Boussinesq equation is a fundamental model in nonlinear wave theory, governing shallow-water wave propagation, coastal dynamics in ocean engineering, and long waves in geophysical fluid systems such as atmospheric and oceanic currents. We present a novel neural network symbolic computation framework that seamlessly integrates neural architectures for powerful function approximation with symbolic manipulation for exact algebraic resolution, eliminating the need for bilinear transformations and thereby substantially reducing computational complexity. Applying this framework, we derive five previously unreported exact analytical solutions using carefully designed neural network configurations and probe functions. These solutions provide valuable tools for modeling ocean internal waves, coastal engineering simulations, and nonlinear optical pulse dynamics. In practice, the method delivers faster and more accurate simulations, improving engineering design and environmental prediction capabilities. By synergistically combining neural networks with symbolic computation, our approach surpasses traditional numerical methods and physics-informed neural networks in both accuracy and efficiency, opening new avenues for solving complex nonlinear partial differential equations.
- 12 February 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (4) , 648
Open AccessA New Neural Network Framework Integrating Symbolic Computation to Solve the (2+1)-Dimensional Boussinesq Equation
The (2+1)-dimensional Boussinesq equation is a fundamental model in nonlinear wave theory, governing shallow-water wave propagation, coastal dynamics in ocean engineering, and long waves in geophysical fluid systems such as atmospheric and oceanic currents. We present a novel neural network symbolic computation framework that seamlessly integrates neural architectures for powerful function approximation with symbolic manipulation for exact algebraic resolution, eliminating the need for bilinear transformations and thereby substantially reducing computational complexity. Applying this framework, we derive five previously unreported exact analytical solutions using carefully designed neural network configurations and probe functions. These solutions provide valuable tools for modeling ocean internal waves, coastal engineering simulations, and nonlinear optical pulse dynamics. In practice, the method delivers faster and more accurate simulations, improving engineering design and environmental prediction capabilities. By synergistically combining neural networks with symbolic computation, our approach surpasses traditional numerical methods and physics-informed neural networks in both accuracy and efficiency, opening new avenues for solving complex nonlinear partial differential equations.
- Open Access
- 30 November 2025
- Published by Academic Publishing Pte. Ltd. in Advances in Differential Equations and Control Processes
- Vol. 32 (4)
Exact solutions of the (2+1)-dimensional BLMP equation via neural network based symbolic computation
This paper introduces a neural network-based symbolic computation framework for deriving exact analytical solutions to nonlinear partial differential equations (NLPDEs). By integrating the expressive capability of neural networks with the interpretability of symbolic methods, the approach offers a flexible, generalizable, and computationally efficient alternative to traditional techniques. Its architecture is straightforward, requiring minimal prior assumptions about the equation structure, thus enabling broad applicability across mathematical physics. As a case study, we investigate the (2+1)-dimensional Boiti-Leon-Manna-Pempinelli (BLMP) equation, a fundamental model describing wave propagation in fluid dynamics, nonlinear optics, and plasma physics. By systematically varying hidden-layer configurations and activation functions, we construct six distinct neural network models—comprising both single- and double-hidden-layer designs. These yield multiple novel exact analytical solutions, revealing diverse dynamic behaviors such as localized wave structures, periodic oscillations, and energy-localized modes. The physical characteristics of the obtained solutions are visualized through three-dimensional surface plots, contour maps, density distributions, and temporal evolution graphs. These illustrations elucidate key features including waveform stability, soliton localization, and nonlinear interactions. Unlike conventional methods such as the Hirota bilinear transformation or inverse scattering, the proposed framework avoids complex algebraic manipulations and does not rely on specialized transformations. Compared to physics-informed neural networks, it achieves higher interpretability and reduced computational dependency. This work expands the solution space of the (2+1)-dimensional BLMP equation and highlights the potential of neural-symbolic computation for discovering exact solutions in complex nonlinear systems.
- 30 November 2025
- Published by Academic Publishing Pte. Ltd. in Advances in Differential Equations and Control Processes
- Vol. 32 (4)
Open AccessExact solutions of the (2+1)-dimensional BLMP equation via neural network based symbolic computation
This paper introduces a neural network-based symbolic computation framework for deriving exact analytical solutions to nonlinear partial differential equations (NLPDEs). By integrating the expressive capability of neural networks with the interpretability of symbolic methods, the approach offers a flexible, generalizable, and computationally efficient alternative to traditional techniques. Its architecture is straightforward, requiring minimal prior assumptions about the equation structure, thus enabling broad applicability across mathematical physics. As a case study, we investigate the (2+1)-dimensional Boiti-Leon-Manna-Pempinelli (BLMP) equation, a fundamental model describing wave propagation in fluid dynamics, nonlinear optics, and plasma physics. By systematically varying hidden-layer configurations and activation functions, we construct six distinct neural network models—comprising both single- and double-hidden-layer designs. These yield multiple novel exact analytical solutions, revealing diverse dynamic behaviors such as localized wave structures, periodic oscillations, and energy-localized modes. The physical characteristics of the obtained solutions are visualized through three-dimensional surface plots, contour maps, density distributions, and temporal evolution graphs. These illustrations elucidate key features including waveform stability, soliton localization, and nonlinear interactions. Unlike conventional methods such as the Hirota bilinear transformation or inverse scattering, the proposed framework avoids complex algebraic manipulations and does not rely on specialized transformations. Compared to physics-informed neural networks, it achieves higher interpretability and reduced computational dependency. This work expands the solution space of the (2+1)-dimensional BLMP equation and highlights the potential of neural-symbolic computation for discovering exact solutions in complex nonlinear systems.
- Open Access
- 27 October 2025
- Published by Springer Nature in Research Square
Bilinear Feature-Enhanced Symbolic Computation Neural Network Method for Solving the Caudrey-Dodd-Gibbon Equation
The fifth-order dispersion nonlinear wave Caudrey-Dodd-Gibbon (CDG) equation is a classic model describing soliton phenomena in fields such as plasma magnetosonic waves and optical fiber light pulses, and its exact solution is of great importance for revealing the laws of nonlinear wave motion. In this paper, an integrated framework of ”bilinear polynomial feature enhancement + symbolic computation constraint + neural network learning” is proposed. Bilinear polynomial features such as x 2 , t 2 , and xt are introduced to break through the input limitation of original variables, broaden the boundary of the model in capturing nonlinear interactions between variables, and reduce errors caused by insufficient feature information. Symbolic computation is applied to the bilinear transformation derivation and conservation law analysis of the CDG equation to provide mathematical structure constraints for the neural network, and a collaborative mechanism of ”symbolic reasoning guiding numerical learning” is constructed to improve the interpretability of the model. This framework breaks down the barriers between traditional numerical and pure neural network methods, realizes efficient and accurate solution of the CDG equation, and provides a new path for the study of exact solutions of high-dimensional, variable-coefficient, and strongly nonlinear partial differential equations.
- 27 October 2025
- Published by Springer Nature in Research Square
Open AccessBilinear Feature-Enhanced Symbolic Computation Neural Network Method for Solving the Caudrey-Dodd-Gibbon Equation
The fifth-order dispersion nonlinear wave Caudrey-Dodd-Gibbon (CDG) equation is a classic model describing soliton phenomena in fields such as plasma magnetosonic waves and optical fiber light pulses, and its exact solution is of great importance for revealing the laws of nonlinear wave motion. In this paper, an integrated framework of ”bilinear polynomial feature enhancement + symbolic computation constraint + neural network learning” is proposed. Bilinear polynomial features such as x 2 , t 2 , and xt are introduced to break through the input limitation of original variables, broaden the boundary of the model in capturing nonlinear interactions between variables, and reduce errors caused by insufficient feature information. Symbolic computation is applied to the bilinear transformation derivation and conservation law analysis of the CDG equation to provide mathematical structure constraints for the neural network, and a collaborative mechanism of ”symbolic reasoning guiding numerical learning” is constructed to improve the interpretability of the model. This framework breaks down the barriers between traditional numerical and pure neural network methods, realizes efficient and accurate solution of the CDG equation, and provides a new path for the study of exact solutions of high-dimensional, variable-coefficient, and strongly nonlinear partial differential equations.
- Open Access
- 17 September 2025
- Published by MDPI AG in Mathematics
- Vol. 13 (18) , 3006
Neural Network-Based Symbolic Computation Algorithm for Solving (2+1)-Dimensional Yu-Toda-Sasa-Fukuyama Equation
This paper presents a Neural Network-Based Symbolic Computation Algorithm (NNSCA) for solving the (2+1)-dimensional Yu-Toda-Sasa-Fukuyama (YTSF) equation. By combining neural networks with symbolic computation, NNSCA bypasses traditional method limitations, deriving and visualizing exact solutions. It designs neural network architectures, converts the PDE into algebraic constraints via Maple, and forms a closed-loop solution process. NNSCA provides a general paradigm for high-dimensional nonlinear PDEs, showing great application potential.
- 17 September 2025
- Published by MDPI AG in Mathematics
- Vol. 13 (18) , 3006
Open AccessNeural Network-Based Symbolic Computation Algorithm for Solving (2+1)-Dimensional Yu-Toda-Sasa-Fukuyama Equation
This paper presents a Neural Network-Based Symbolic Computation Algorithm (NNSCA) for solving the (2+1)-dimensional Yu-Toda-Sasa-Fukuyama (YTSF) equation. By combining neural networks with symbolic computation, NNSCA bypasses traditional method limitations, deriving and visualizing exact solutions. It designs neural network architectures, converts the PDE into algebraic constraints via Maple, and forms a closed-loop solution process. NNSCA provides a general paradigm for high-dimensional nonlinear PDEs, showing great application potential.