Summary of A Survey Of Few-shot Learning on Graphs: From Meta-learning to Pre-training and Prompt Learning, by Xingtong Yu et al.
A Survey of Few-Shot Learning on Graphs: from Meta-Learning to Pre-Training and Prompt Learning
by Xingtong Yu, Yuan Fang, Zemin Liu, Yuxia Wu, Zhihao Wen, Jianyuan Bo, Xinming Zhang, Steven C.H. Hoi
First submitted to arxiv on: 2 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Social and Information Networks (cs.SI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper surveys the recent developments in few-shot learning on graphs, a crucial step in graph-centric tasks. The authors provide a comprehensive overview of existing studies, categorizing them into two major taxonomies: problem taxonomy and technique taxonomy. The techniques include meta-learning, pre-training, and hybrid approaches, with comparisons of their strengths and limitations. This study aims to synthesize recent advancements, identify future directions, and aid readers in selecting the most suitable method for their applications. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about how computers can learn new things from small amounts of data on graphs, which are like maps that connect different pieces of information. Right now, it’s hard for computers to learn from very little data, so this study looks at what has been done already and tries to figure out what we need to do next. It groups all the previous work into two main categories: one is about what kind of problem we’re trying to solve, and the other is about how we can use different techniques to solve those problems. |
Keywords
* Artificial intelligence * Few shot * Meta learning
