Abstract
We propose Graph-Coupled Oscillator Networks (GraphCON), a novel framework for deep learning on graphs. It is based on discretizations of a second-order system of ordinary differential equations (ODEs), which model a network of nonlinear forced and damped oscillators, coupled via the adjacency structure of the underlying graph. The flexibility of our framework permits any basic GNN layer (e.g. convolutional or attentional) as the coupling function, from which a multi-layer deep neural network is built up via the dynamics of the proposed ODEs. We relate the oversmoothing problem, commonly encountered in GNNs, to the stability of steady states of the underlying ODE and show that zero-Dirichlet energy steady states are not stable for our proposed ODEs. This demonstrates that the proposed framework mitigates the oversmoothing problem. Finally, we show that our approach offers competitive performance with respect to the state-of-the-art on a variety of graph-based learning tasks. Show more
Publication status
publishedExternal links
Journal / series
SAM Research ReportVolume
Publisher
Seminar for Applied Mathematics, ETH ZurichOrganisational unit
03851 - Mishra, Siddhartha / Mishra, Siddhartha
Funding
770880 - Computation and analysis of statistical solutions of fluid flow (EC)
Related publications and datasets
Is previous version of: http://hdl.handle.net/20.500.11850/610760
More
Show all metadata
ETH Bibliography
yes
Altmetrics