Poster
in
Workshop: Neural Network Weights as a New Data Modality
Fusion of Graph Neural Networks via Optimal Transport
Weronika Ormaniec · Michael Vollenweider · Elisa Hoskovec
Keywords: [ Graph Convolutional Networks ] [ Fused Gromov-Wasserstein ] [ Model Fusion ] [ Ensemble ] [ Optimal Transport ] [ Wasserstein Barycenter ] [ Graph Neural Networks ] [ ConFusion ]
In this paper, we explore the idea of combining multiple Graph Convolutional Networks (GCNs) into one model. To that end, we align the weights of different models layer-wise using optimal transport (OT). We present and evaluate three types of transportation costs and show that the studied fusion method consistently outperforms the performance of vanilla averaging. Finally, we present results suggesting that model fusion using OT is harder in the case of GCNs than MLPs and that incorporating the graph structure into the process does not improve the performance of the method.