The paper reviews the development and techniques of few-shot learning on graphs, a field that has seen significant advancements in recent years. The authors categorize existing studies into three major families: meta-learning approaches, pre-training approaches, and hybrid approaches. Each category’s strengths and limitations are compared, and the relationships among these methods are analyzed. The paper aims to provide a systematic review of prevailing methods for few-shot learning on graphs and to identify future research avenues.
Publication date: 5 Feb 2024
Project Page: Not provided
Paper: https://arxiv.org/pdf/2402.01440
