VIGraph: Generative Self-supervised Learning for Class-Imbalanced Node Classification

TL;DR

VIGraph is a novel generative self-supervised learning method that effectively addresses class imbalance in node classification by generating high-quality minority class nodes using Variational GAE, outperforming existing approaches.

Contribution

Introducing VIGraph, a generative SSL approach utilizing Variational GAE and innovative training strategies to improve class-imbalanced node classification without retraining.

Findings

VIGraph outperforms SMOTE-based methods in imbalanced node classification.

VIGraph generates high-quality minority nodes directly usable for classification.

Extensive experiments demonstrate the effectiveness and generality of VIGraph.

Abstract

Class imbalance in graph data presents significant challenges for node classification. While existing methods, such as SMOTE-based approaches, partially mitigate this issue, they still exhibit limitations in constructing imbalanced graphs. Generative self-supervised learning (SSL) methods, exemplified by graph autoencoders (GAEs), offer a promising solution by directly generating minority nodes from the data itself, yet their potential remains underexplored. In this paper, we delve into the shortcomings of SMOTE-based approaches in the construction of imbalanced graphs. Furthermore, we introduce VIGraph, a simple yet effective generative SSL approach that relies on the Variational GAE as the fundamental model. VIGraph strictly adheres to the concept of imbalance when constructing imbalanced graphs and innovatively leverages the variational inference (VI) ability of Variational GAE to generate nodes for minority classes. VIGraph introduces comprehensive training strategies, including cross-view contrastive learning at the decoding phase to capture semantic knowledge, adjacency matrix reconstruction to preserve graph structure, and alignment strategy to ensure stable training. VIGraph can generate high-quality nodes directly usable for classification, eliminating the need to integrate the generated nodes back to the graph as well as additional retraining found in SMOTE-based methods. We conduct extensive experiments, results from which demonstrate the superiority and generality of our approach.