研究方向

研究聚焦于科学机器学习, 包括算子学习, 保结构神经网络, 混合算法等.

工作经历

2024.6-至今: 北京大学, 大数据分析与应用技术国家工程实验室, 助理研究员

2021.7-2024.5: 北京大学, 数学科学学院, 博雅博士后

2019.3-2019.6&2020.1-2021.1: Brown University, Division of Applied Mathematics, 访问学者

教育经历

2016.9-2021.7: 中国科学院数学与系统科学研究院, 计算数学所, 理学博士

2012.9-2016.7: 中国科学技术大学, 数学科学学院, 理学学士

科研项目

3. 国家重点研发计划青年科学家项目, 神经算子学习: 理论、方法及应用, 2025.12-2030.11, 骨干.

2. 国家自然科学基金青年科学基金项目(C类), 变区域偏微分方程算子学习方法, 2026.1-2028.12, 主持.

1. 中国博士后科学基金第71批面上资助一等, 保结构机器学习方法研究, 2022.6-2024.5, 主持.

已发表论文 (未标明通讯的文章按姓氏排序)

14. Shanshan Xiao, Pengzhan Jin*, Yifa Tang. A Deformation-Based Framework for Learning Solution Mappings of PDEs Defined on Varying Domains. SIAM Journal on Numerical Analysis, 64(2), 537-564, 2026.

13. Pengzhan Jin. Two-hidden-layer ReLU neural networks and finite elements. Neural Networks, 198, 108559, 2026.

12. Jun Hu, Pengzhan Jin. A hybrid iterative method based on MIONet for PDEs: Theory and numerical examples. Mathematics of Computation, 95(359), 1327-1359, 2026.

11. Yifan Wang, Hehu Xie*, Pengzhan Jin. Tensor Neural Network and Its Numerical Integration. Journal of Computational Mathematics, 42(6), 1714-1742, 2024.

10. Pengzhan Jin, Zhen Zhang, Ioannis G. Kevrekidis, George Em Karniadakis*. Learning Poisson systems and trajectories of autonomous systems via Poisson neural networks. IEEE Transactions on Neural Networks and Learning Systems, 34(11), 8271-8283, 2023.

9. Pengzhan Jin, Shuai Meng, Lu Lu*. MIONet: Learning multiple-input operators via tensor product. SIAM Journal on Scientific Computing, 44(6), A3490-A3514, 2022.

8. Aiqing Zhu, Pengzhan Jin, Beibei Zhu, Yifa Tang*. On Numerical Integration in Neural Ordinary Differential Equations. Proceedings of the 39th International Conference on Machine Learning, PMLR 162, 27527-27547, 2022.

7. Pengzhan Jin, Zhangli Lin, Bo Xiao. Optimal unit triangular factorization of symplectic matrices. Linear Algebra and its Applications, 650, 236–247, 2022.

6. Aiqing Zhu, Pengzhan Jin, Yifa Tang*. Approximation capabilities of measure-preserving neural networks. Neural Networks, 147, 72-80, 2021.

5. Lu Lu, Pengzhan Jin, Guofei Pang, Zhongqiang Zhang, George Em Karniadakis*. Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators. Nature Machine Intelligence, 3, 218–229, 2021.

4. Aiqing Zhu, Pengzhan Jin, Yifa Tang*. Deep Hamiltonian neural networks based on symplectic integrators (in Chinese). Mathematica Numerica Sinica, 42(3), 370-384, 2020.

3. Pengzhan Jin, Zhen Zhang, Aiqing Zhu, Yifa Tang*, George Em Karniadakis*. SympNets: Intrinsic structure-preserving symplectic networks for identifying Hamiltonian systems. Neural Networks, 132, 166-179, 2020.

2. Pengzhan Jin, Yifa Tang, Aiqing Zhu. Unit triangular factorization of the matrix symplectic group. SIAM Journal on Matrix Analysis and Applications, 41(4), 1630-1650, 2020.

1. Pengzhan Jin, Lu Lu, Yifa Tang, George Em Karniadakis*. Quantifying the generalization error in deep learning in terms of data distribution and neural network smoothness. Neural Networks, 130, 85-99, 2020.

预印本

4. Jun Hu, Pengzhan Jin, Weijun Zhang. Manifold Function Encoder: Identifying Different Functions Defined on Different Manifolds. arXiv preprint arXiv:2512.20227, 2025.

3. Shanshan Xiao, Pengzhan Jin*, Yifa Tang. Learning solution operators of PDEs defined on varying domains via MIONet. arXiv preprint arXiv:2402.15097, 2024.

2. Jun Hu, Pengzhan Jin. Experimental observation on a low-rank tensor model for eigenvalue problems. arXiv preprint arXiv:2302.00538, 2023.

1. Lu Lu, Pengzhan Jin, George Em Karniadakis*. DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators. arXiv preprint arXiv:1910.03193, 2019.