{"slug": "learning-coarse-step-dynamics-and-internal-mechanical-response-with-graph", "title": "Learning coarse-step dynamics and internal mechanical response with graph networks", "summary": "Researchers introduced Newmark-β-DGN, a graph neural network framework that combines a semi-implicit Newmark-β-inspired update with an operator-weighted virtual hub to predict coarse-step physical dynamics and infer unobserved mechanical quantities, according to an arXiv paper (arXiv:2609.30344v1). Tested on a deformable beam, human motion and protein dynamics, the framework supported long-horizon prediction at time steps where explicit learned simulators deteriorate. Without force, moment or constitutive relation supervision, forces inferred from walking kinematics tracked independently derived hip and knee joint moments, and response operators learned on the beam recovered the relative spatial and directional structure of its finite-element stiffness tangent.", "body_md": "arXiv:2609.30344v1 Announce Type: new \nAbstract: Modern sensing records the motion of physical systems, but often leaves the forces and mechanical response governing that motion unobserved. Inferring these quantities from discretely sampled trajectories is especially difficult at coarse time scales, when mechanical response evolves between observations and interactions propagate across the system. Here we introduce Newmark-\\b{eta}-DGN, a graph neural network-based framework that combines two structures inspired by computational mechanics. First, a semi-implicit update inspired by the Newmark-\\b{eta} method uses learned momentum fluxes and matrix-valued response operators to advance the state over each observed interval. Second, an operator-weighted virtual hub provides system-wide coupling through a sparse set of connections. The learned quantities thus determine the predicted motion and remain accessible for mechanical analysis. Across a deformable beam, human motion and protein dynamics, Newmark-\\b{eta}-DGN supports long-horizon prediction at time steps for which explicit learned simulators deteriorate. Without force, moment or constitutive relation supervision, forces inferred from walking kinematics track independently derived hip and knee joint moments, while response operators learned on the beam recover the relative spatial and directional structure of its finite-element stiffness tangent. Newmark-\\b{eta}-DGN therefore links coarse-step prediction to the inference of mechanical quantities that were never observed during training.", "url": "https://wpnews.pro/news/learning-coarse-step-dynamics-and-internal-mechanical-response-with-graph", "canonical_source": "https://arxiv.org/abs/2609.30344", "published_at": "2026-09-29 04:00:00+00:00", "updated_at": "2026-09-29 04:18:21.721859+00:00", "lang": "en", "topics": ["machine-learning", "neural-networks", "ai-research"], "entities": ["Newmark-β-DGN", "arXiv"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/learning-coarse-step-dynamics-and-internal-mechanical-response-with-graph", "markdown": "https://wpnews.pro/news/learning-coarse-step-dynamics-and-internal-mechanical-response-with-graph.md", "text": "https://wpnews.pro/news/learning-coarse-step-dynamics-and-internal-mechanical-response-with-graph.txt", "jsonld": "https://wpnews.pro/news/learning-coarse-step-dynamics-and-internal-mechanical-response-with-graph.jsonld"}}