Jul 2026
Learning Networked Dynamic System Models with Weak Form and Graph Neural Networks
The weak graph Koopman bilinear form model is proposed, which integrates geometric deep learning and Koopman theory to learn latent-space dynamics for networked systems, especially for challenging cases that have multiple timescales.
Yin Yu, Daning Huang, Seho Park et al.
· Journal of Guidance Control... · 0 citations