Template based Graph Neural Network with Optimal Transport Distances - Archive ouverte HAL Access content directly
Conference Papers Year : 2022

Template based Graph Neural Network with Optimal Transport Distances

Abstract

Current Graph Neural Networks (GNN) architectures generally rely on two important components: node features embedding through message passing, and aggregation with a specialized form of pooling. The structural (or topological) information is implicitly taken into account in these two steps. We propose in this work a novel point of view, which places distances to some learnable graph templates at the core of the graph representation. This distance embedding is constructed thanks to an optimal transport distance: the Fused Gromov-Wasserstein (FGW) distance, which encodes simultaneously feature and structure dissimilarities by solving a soft graph-matching problem. We postulate that the vector of FGW distances to a set of template graphs has a strong discriminative power, which is then fed to a non-linear classifier for final predictions. Distance embedding can be seen as a new layer, and can leverage on existing message passing techniques to promote sensible feature representations. Interestingly enough, in our work the optimal set of template graphs is also learnt in an end-to-end fashion by differentiating through this layer. After describing the corresponding learning procedure, we empirically validate our claim on several synthetic and real life graph classification datasets, where our method is competitive or surpasses kernel and GNN state-of-the-art approaches. We complete our experiments by an ablation study and a sensitivity analysis to parameters.
Fichier principal
Vignette du fichier
FGWM.pdf (1.49 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03839517 , version 1 (04-11-2022)

Identifiers

  • HAL Id : hal-03839517 , version 1

Cite

Cédric Vincent-Cuaz, Rémi Flamary, Marco Corneli, Titouan Vayer, Nicolas Courty. Template based Graph Neural Network with Optimal Transport Distances. NeurIPS 2022 – 36th Conference on Neural Information Processing Systems, Nov 2022, New Orleans, United States. ⟨hal-03839517⟩
157 View
137 Download

Share

Gmail Mastodon Facebook X LinkedIn More