Graph Mamba Operator: A Latent Simulator for Interacting Particle Systems
2026-06-08 • Machine Learning
Machine Learning
AI summaryⓘ
The authors developed a new method called GraMO to better predict how complex systems change over time while considering both space and time together. Unlike older methods that treat space and time separately and often make mistakes when predicting far into the future, GraMO combines these factors in one step using a special mathematical update. They tested GraMO on different tasks like simulating physical objects and robot movements, where it made fewer errors than other methods, especially for long-term predictions. This shows their approach can better understand interactions that happen over space and time simultaneously.
graph neural networksdynamical systemsstate-space modelslatent-space simulationautoregressive modelslong-range dependenciesN-body problemmotion capturerobotics datasetstemporal dynamics
Authors
Karn Tiwari, Niladri Dutta, N M Anoop Krishnan, Prathosh A P
Abstract
Modeling interacting dynamical systems requires capturing spatial interactions alongside long-range temporal dependencies. Graph neural networks (GNNs) provide a natural representation but typically rely on autoregressive rollouts and treat spatial and temporal dynamics separately, leading to error accumulation over long horizons. Existing approaches also focus on local interactions and short temporal contexts, limiting their ability to capture multi-hop dependencies and global structure. We introduce the Graph Mamba Operator (GraMO), a latent-space simulator that integrates state-space models with graph-based interaction learning. In contrast to prior work that sequences nodes or applies spatial and temporal updates in separate stages, GraMO couples graph-based interactions and temporal state updates within a single recurrence. The update is linear in the latent state, with input-dependent coefficients that adapt across regimes. We evaluate GraMO on N-body systems, motion capture, and robotics datasets, achieving the lowest error across benchmarks and the largest gains in long-horizon prediction.