Reduced-Order Modeling of Hamiltonian Dynamics Based on Symplectic Neural Networks

Abstract. We introduce a novel data-driven symplectic reduced-order modeling (ROM) framework for high-dimensional Hamiltonian systems that unifies latent-space discovery and dynamics learning within a single, end-to-end neural architecture. The encoder–decoder is built from Hénon neural networks (HénonNets) and may be augmented with linear G-reflector layers, yielding a structure-preserving symplectic embedding and an associated one-sided symplectic pseudo-inverse between the full and latent phase spaces. Latent dynamics are advanced by a symplectic flow map implemented as a HénonNet. This unified neural architecture ensures exact preservation of the underlying symplectic structure at the reduced-order level, significantly enhancing the fidelity and long-term stability of the resulting ROM. We validate our method through comprehensive numerical experiments on canonical Hamiltonian systems. The results demonstrate the method’s capability for accurate trajectory reconstruction, robust predictive performance beyond the training horizon and strong Hamiltonian preservation. These promising outcomes underscore the effectiveness and potential applicability of our symplectic ROM framework for complex dynamical systems across a broad range of scientific and engineering disciplines. The implementation associated with this work is publicly available at https://github.com/chenyongssss/Symp_NN .