Enhancing Unsupervised Graph Few-shot Learning via Set Functions and Optimal Transport

TL;DR

This paper introduces STAR, a novel approach that combines set functions and optimal transport to improve unsupervised graph few-shot learning by capturing set-level features and aligning support and query distributions.

Contribution

STAR is the first model to integrate set functions and optimal transport for unsupervised graph few-shot learning, addressing distribution shift and limited labeled data issues.

Findings

STAR outperforms existing methods on multiple datasets.

Theoretical analysis confirms improved task relevance and generalization.

Empirical results demonstrate enhanced performance in real-world scenarios.

Abstract

Graph few-shot learning has garnered significant attention for its ability to rapidly adapt to downstream tasks with limited labeled data, sparking considerable interest among researchers. Recent advancements in graph few-shot learning models have exhibited superior performance across diverse applications. Despite their successes, several limitations still exist. First, existing models in the meta-training phase predominantly focus on instance-level features within tasks, neglecting crucial set-level features essential for distinguishing between different categories. Second, these models often utilize query sets directly on classifiers trained with support sets containing only a few labeled examples, overlooking potential distribution shifts between these sets and leading to suboptimal performance. Finally, previous models typically require necessitate abundant labeled data from base classes to extract transferable knowledge, which is typically infeasible in real-world scenarios. To address these issues, we propose a novel model named STAR, which leverages Set funcTions and optimAl tRansport for enhancing unsupervised graph few-shot learning. Specifically, STAR utilizes expressive set functions to obtain set-level features in an unsupervised manner and employs optimal transport principles to align the distributions of support and query sets, thereby mitigating distribution shift effects. Theoretical analysis demonstrates that STAR can capture more task-relevant information and enhance generalization capabilities. Empirically, extensive experiments across multiple datasets validate the effectiveness of STAR. Our code can be found here.