Neural Ordinary Differential Equation Control of Dynamics on Graphs

Autor:	Asikis, Thomas, Böttcher, Lucas, Antulov-Fantulin, Nino
Rok vydání:	2020
Předmět:	Computer Science - Machine Learning Computer Science - Social and Information Networks Statistics - Machine Learning
Zdroj:	Phys. Rev. Research 4, 013221 (2022)
Druh dokumentu:	Working Paper
DOI:	10.1103/PhysRevResearch.4.013221
Popis:	We study the ability of neural networks to calculate feedback control signals that steer trajectories of continuous time non-linear dynamical systems on graphs, which we represent with neural ordinary differential equations (neural ODEs). To do so, we present a neural-ODE control (NODEC) framework and find that it can learn feedback control signals that drive graph dynamical systems into desired target states. While we use loss functions that do not constrain the control energy, our results show, in accordance with related work, that NODEC produces low energy control signals. Finally, we evaluate the performance and versatility of NODEC against well-known feedback controllers and deep reinforcement learning. We use NODEC to generate feedback controls for systems of more than one thousand coupled, non-linear ODEs that represent epidemic processes and coupled oscillators. Comment: Fifth version improves and clears notation
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2006.09773 Zobrazit plný text záznamu View this record from Arxiv