Label-GCN: an effective method for adding label propagation to graph convolutional networks - Télécom SudParis Access content directly
Preprints, Working Papers, ... Year : 2021

Label-GCN: an effective method for adding label propagation to graph convolutional networks

Abstract

We show that a modification of the first layer of a Graph Convolutional Network (GCN) can be used to effectively propagate label information across neighbor nodes, for binary and multi-class classification problems. This is done by selectively eliminating self-loops for the label features during the training phase of a GCN. The GCN architecture is otherwise unchanged, without any extra hyper-parameters, and can be used in both a transductive and inductive setting. We show through several experiments that, depending on how many labels are available during the inference phase, this strategy can lead to a substantial improvement in the model performance compared to a standard GCN approach, including with imbalanced datasets.
Fichier principal
Vignette du fichier
2104.02153.pdf (528.41 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03991083 , version 1 (18-06-2024)

Identifiers

Cite

Claudio Bellei, Hussain Alattas, Nesrine Kaaniche. Label-GCN: an effective method for adding label propagation to graph convolutional networks. 2024. ⟨hal-03991083⟩
28 View
3 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More