<ul><li>Accurately predicting long-horizon molecular dynamics (MD) trajectories remains a significant challenge, as existing deep learning methods often struggle to retain fidelity over extended simulations.</li><li>To address these limitations, the researchers propose Graph Fourier Neural ODEs (GF-NODE), which integrates a graph Fourier transform for spatial frequency decomposition with a Neural ODE framework for continuous-time evolution.</li><li>GF-NODE decomposes molecular configurations into multiple spatial frequency modes, evolves the frequency components in time, and reconstructs the updated molecular geometry through an inverse graph Fourier transform.</li><li>Experimental results on challenging MD benchmarks demonstrate that GF-NODE achieves state-of-the-art accuracy while preserving essential geometrical features over extended simulations.</li></ul>

Graph Fourier Neural ODEs: Modeling Spatial-temporal Multi-scales in Molecular Dynamics

Discover more