Skip to main content
QUICK REVIEW

[论文解读] Graph-based Semi-supervised Learning: A Comprehensive Review

Zixing Song, Xiangli Yang|arXiv (Cornell University)|Feb 26, 2021
Text and Document Classification Technologies被引用 26
一句话总结

本综述提供了系统且最新的图基半监督学习(GSSL)分类学和批判性评述,详细介绍图构建与标签推断方法,并以统一的编码–解码视角讨论图嵌入,以及对未来方向的讨论。

ABSTRACT

Semi-supervised learning (SSL) has tremendous value in practice due to its ability to utilize both labeled data and unlabelled data. An important class of SSL methods is to naturally represent data as graphs such that the label information of unlabelled samples can be inferred from the graphs, which corresponds to graph-based semi-supervised learning (GSSL) methods. GSSL methods have demonstrated their advantages in various domains due to their uniqueness of structure, the universality of applications, and their scalability to large scale data. Focusing on this class of methods, this work aims to provide both researchers and practitioners with a solid and systematic understanding of relevant advances as well as the underlying connections among them. This makes our paper distinct from recent surveys that cover an overall picture of SSL methods while neglecting fundamental understanding of GSSL methods. In particular, a major contribution of this paper lies in a new generalized taxonomy for GSSL, including graph regularization and graph embedding methods, with the most up-to-date references and useful resources such as codes, datasets, and applications. Furthermore, we present several potential research directions as future work with insights into this rapidly growing field.

研究动机与目标

  • 提供对GSSL进展的扎实、系统性理解以及它们之间的关系。
  • 提出将图构建与标签推断分离的通用分类法。
  • 在统一框架内总结图正则化和图嵌入方法。
  • 为研究人员和从业者提供实用资源,包括代码、数据集和应用。
  • 识别动态性、可扩展性、噪声鲁棒性与稳健性方面的开放问题与未来方向。

提出的方法

  • 引入一个两步的GSSL过程:图构建与标签推断。
  • 给出广义的损失框架 L(f) = Ls(f, Dl) + μLr(f, D) 并将其具体化为图正则化与图嵌入。
  • 将标签推断方法分为图正则化与图嵌入;讨论具体的正则化项及其形式。
  • 提出一个统一的编码–解码视角用于图嵌入,以连接浅层与深层嵌入方法。
  • 回顾具体方法:标签传播、GRF、LGC、p-Laplacian正则化、有向正则化、流形正则化、LPDGL、Poisson学习。
  • 讨论诸如KNN、ε-邻域和b-匹配等图构建技术,以及有监督变体。

实验结果

研究问题

  • RQ1在GSSL中,图构建与标签推断如何系统化地分类?
  • RQ2哪些核心的正则化和嵌入方法能够实现对图上有效的标签传播?
  • RQ3如何在编码–解码框架下统一图嵌入,以同时捕捉浅层与深层表示?
  • RQ4哪些资源(代码、数据集、应用)支持GSSL的实际应用与试验?
  • RQ5在GSSL的动态性、可扩展性、噪声鲁棒性与稳健性方面有哪些开放研究方向?

主要发现

  • 本文提供了对GSSL的全面、最新的综述与一个新的分类法。
  • 它提出一个一般化的损失框架,指导图正则化与图嵌入方法。
  • 标签传播及其变体(GRF、LGC、p-Laplacian、有向正则化)是图正则化的核心。
  • 在编码–解码范式下,将图嵌入置于统一框架,连接浅层与深层方法。
  • 该综述整理了可用的代码、数据集与应用,支持实际试验。
  • 它讨论了在动态、可扩展和鲁棒的GSSL方面的开放问题与未来方向。

更好的研究,从现在开始

从阅读论文到最终审阅,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。