Graph Contrastive Learning Meets Graph Meta Learning: A Unified Method for Few-shot Node Tasks

TL;DR

This paper introduces COLA, a unified framework combining contrastive and meta learning for few-shot node classification, leveraging graph augmentations to improve generalization and achieve state-of-the-art results.

Contribution

The paper proposes COLA, a novel paradigm that integrates contrastive learning with meta learning for few-shot node tasks, utilizing graph augmentations to enhance performance.

Findings

COLA outperforms existing methods on all tested tasks.

Graph augmentations effectively identify semantically similar nodes.

Contrastive learning's advantages include comprehensive node utilization and augmentation power.

Abstract

Graph Neural Networks (GNNs) have become popular in Graph Representation Learning (GRL). One fundamental application is few-shot node classification. Most existing methods follow the meta learning paradigm, showing the ability of fast generalization to few-shot tasks. However, recent works indicate that graph contrastive learning combined with fine-tuning can significantly outperform meta learning methods. Despite the empirical success, there is limited understanding of the reasons behind it. In our study, we first identify two crucial advantages of contrastive learning compared to meta learning, including (1) the comprehensive utilization of graph nodes and (2) the power of graph augmentations. To integrate the strength of both contrastive learning and meta learning on the few-shot node classification tasks, we introduce a new paradigm: Contrastive Few-Shot Node Classification (COLA). Specifically, COLA employs graph augmentations to identify semantically similar nodes, which enables the construction of meta-tasks without the need for label information. Therefore, COLA can utilize all nodes to construct meta-tasks, further reducing the risk of overfitting. Through extensive experiments, we validate the essentiality of each component in our design and demonstrate that COLA achieves new state-of-the-art on all tasks.