Improving Long-Range Interactions in Graph Neural Simulators via Hamiltonian Dynamics
This work addresses a key bottleneck in data-driven physical simulation for applications requiring long-range dependencies, though it appears incremental by building on existing GNS methods.
The authors tackled the problem of Graph Neural Simulators struggling with long-range interactions and error accumulation by proposing IGNS, which uses Hamiltonian dynamics to preserve information and handle broader dynamics, resulting in higher accuracy and stability across new benchmarks.
Learning to simulate complex physical systems from data has emerged as a promising way to overcome the limitations of traditional numerical solvers, which often require prohibitive computational costs for high-fidelity solutions. Recent Graph Neural Simulators (GNSs) accelerate simulations by learning dynamics on graph-structured data, yet often struggle to capture long-range interactions and suffer from error accumulation under autoregressive rollouts. To address these challenges, we propose Information-preserving Graph Neural Simulators (IGNS), a graph-based neural simulator built on the principles of Hamiltonian dynamics. This structure guarantees preservation of information across the graph, while extending to port-Hamiltonian systems allows the model to capture a broader class of dynamics, including non-conservative effects. IGNS further incorporates a warmup phase to initialize global context, geometric encoding to handle irregular meshes, and a multi-step training objective to reduce rollout error. To evaluate these properties systematically, we introduce new benchmarks that target long-range dependencies and challenging external forcing scenarios. Across all tasks, IGNS consistently outperforms state-of-the-art GNSs, achieving higher accuracy and stability under challenging and complex dynamical systems.