SGR: Self-Supervised Spectral Graph Representation Learning
Anton Tsitsulin Hasso Plattner InstitutePotsdamGermany , Davide Mottin Hasso Plattner InstitutePotsdamGermany , Panagiotis Karras Aarhus UniversityAarhusDenmark , Alexander Bronstein TechnionHaifaIsrael and Emmanuel Müller Hasso Plattner InstitutePotsdamGermany
Abstract
Representing a graph as a vector is a challenging task; ideally, the representation should be easily computable and conducive to efficient comparisons among graphs, tailored to the particular data and analytical task at hand. Unfortunately, a “one-size-fits-all” solution is unattainable, as different analytical tasks may require different attention to global or local graph features. We develop SGR, the first, to our knowledge, method for learning graph representations in a self-supervised manner. Grounded on spectral graph analysis, SGR seamlessly combines all aforementioned desirable properties. In extensive experiments, we show how our approach works on large graph collections, facilitates self-supervised representation learning across a variety of application domains, and performs competitively to state-of-the-art methods without re-training.
中文速览
把图（graph）表示成向量一直是个难题：既要计算高效，又要能同时捕捉节点连接这类局部特征和社区结构这类全局特征，还得不依赖人工标注。SGR（Spectral Graph Representation）提出用图的拉普拉斯谱（Laplacian spectrum）作为输入，通过一个单层感知机把谱信息映射成固定维度的向量表示，而这个感知机的训练完全不需要真实标注数据——它只需在合成图上学会区分"随机图"和"具有社区结构的随机块模型图"，因此属于自监督学习。在多个真实数据集上的图分类实验中，SGR无需针对新数据集重新训练，就能与当前最优方法持平甚至更优，同时计算开销远低于核方法和监督神经网络。这项工作的重要性在于，它首次将自监督学习引入图级别的表示学习，打破了"要么需要标注、要么无法自适应"的两难困境，为大规模图集合的分析提供了一种灵活、轻量的新范式。
原文 arXiv:1811.06237;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1811.06237v1