iN2V: Bringing Transductive Node Embeddings to Inductive Graphs

TL;DR

The paper introduces iN2V, a method that extends transductive node embeddings like node2vec to inductive settings, enabling effective embedding of unseen nodes and improving node classification performance.

Contribution

iN2V is a novel plug-in approach that adapts node2vec for inductive learning, allowing embeddings for unseen nodes and enhancing existing embedding methods.

Findings

iN2V improves node classification accuracy by 1 point on average.

Up to 6 points of improvement depending on dataset and unseen nodes.

The method is versatile and can be combined with other embedding techniques.

Abstract

Shallow node embeddings like node2vec (N2V) can be used for nodes without features or to supplement existing features with structure-based information. Embedding methods like N2V are limited in their application on new nodes, which restricts them to the transductive setting where the entire graph, including the test nodes, is available during training. We propose inductive node2vec (iN2V), which combines a post-hoc procedure to compute embeddings for nodes unseen during training and modifications to the original N2V training procedure to prepare the embeddings for this post-hoc procedure. We conduct experiments on several benchmark datasets and demonstrate that iN2V is an effective approach to bringing transductive embeddings to an inductive setting. Using iN2V embeddings improves node classification by 1 point on average, with up to 6 points of improvement depending on the dataset and the number of unseen nodes. Our iN2V is a plug-in approach to create new or enrich existing embeddings. It can also be combined with other embedding methods, making it a versatile approach for inductive node representation learning. Code to reproduce the results is available at https://github.com/Foisunt/iN2V .