Virtual Node Tuning for Few-shot Node Classification

TL;DR

This paper introduces Virtual Node Tuning, a novel method for few-shot node classification that uses virtual nodes as soft prompts in a pretrained graph transformer, effectively handling scenarios with limited labeled data.

Contribution

The paper proposes Virtual Node Tuning (VNT), a new approach that employs virtual nodes and a pseudo prompt evolution module to improve few-shot node classification, especially with sparse labels.

Findings

VNT outperforms state-of-the-art methods on four datasets.

VNT effectively handles unlabeled or sparsely labeled base classes.

VNT surpasses fully supervised baselines in few-shot scenarios.

Abstract

Few-shot Node Classification (FSNC) is a challenge in graph representation learning where only a few labeled nodes per class are available for training. To tackle this issue, meta-learning has been proposed to transfer structural knowledge from base classes with abundant labels to target novel classes. However, existing solutions become ineffective or inapplicable when base classes have no or limited labeled nodes. To address this challenge, we propose an innovative method dubbed Virtual Node Tuning (VNT). Our approach utilizes a pretrained graph transformer as the encoder and injects virtual nodes as soft prompts in the embedding space, which can be optimized with few-shot labels in novel classes to modulate node embeddings for each specific FSNC task. A unique feature of VNT is that, by incorporating a Graph-based Pseudo Prompt Evolution (GPPE) module, VNT-GPPE can handle scenarios with sparse labels in base classes. Experimental results on four datasets demonstrate the superiority of the proposed approach in addressing FSNC with unlabeled or sparsely labeled base classes, outperforming existing state-of-the-art methods and even fully supervised baselines.