A Comprehensive Survey of Few-shot Information Networks  

作　　者：Xinxin Zheng Feihu Che Jianhua Tao 

机构地区：[1]School of Artificial Intelligence,University of Chinese Academy of Sciences,Beijing,100049,China [2]Institute of Automation,Chinese Academy of Sciences,Beijing,100190,China [3]Department of Automation,Tsinghua University,Beijing,100084,China

出　　处：《Machine Intelligence Research》2025年第1期60-78,共19页机器智能研究(英文版)

摘　　要：Information networks store rich information in the nodes and edges,which benefit many downstream tasks,such as recommender systems and knowledge graph completion.Information networks contain homogeneous information,heterogeneous information and knowledge graphs.A significant number of surveys focus on one of the three parts and summarize the research works,but few surveys conclude and compare the three kinds of information networks.In addition,in real scenarios,lots of information networks lack sufficient labeled data,so the combination of meta-learning and information networks can bring in extended applications.This paper concentrates on few-shot information networks and systematically presents recent works to help analyze and follow related works.

关 键 词：Few-shot learning META-LEARNING homogeneous information networks heterogeneous information networks knowledge graphs 

分 类 号：TN9[电子电信—信息与通信工程]