Graph Neural Networks: Methods, Applications, and Opportunities

TL;DR

This survey comprehensively reviews graph neural networks (GNNs), covering their methods, applications, and challenges across various learning paradigms, highlighting recent advances and future opportunities in non-Euclidean data analysis.

Contribution

It offers a detailed taxonomy of GNN methods across learning settings, analyzes approaches from theoretical and empirical perspectives, and provides architecture guidelines and open challenges.

Findings

GNNs are effective across supervised, unsupervised, semi-supervised, and self-supervised tasks.

The survey identifies key applications and benchmark datasets for GNNs.

Open challenges include scalability and generalization of GNNs.

Abstract

In the last decade or so, we have witnessed deep learning reinvigorating the machine learning field. It has solved many problems in the domains of computer vision, speech recognition, natural language processing, and various other tasks with state-of-the-art performance. The data is generally represented in the Euclidean space in these domains. Various other domains conform to non-Euclidean space, for which graph is an ideal representation. Graphs are suitable for representing the dependencies and interrelationships between various entities. Traditionally, handcrafted features for graphs are incapable of providing the necessary inference for various tasks from this complex data representation. Recently, there is an emergence of employing various advances in deep learning to graph data-based tasks. This article provides a comprehensive survey of graph neural networks (GNNs) in each learning setting: supervised, unsupervised, semi-supervised, and self-supervised learning. Taxonomy of each graph based learning setting is provided with logical divisions of methods falling in the given learning setting. The approaches for each learning task are analyzed from both theoretical as well as empirical standpoints. Further, we provide general architecture guidelines for building GNNs. Various applications and benchmark datasets are also provided, along with open challenges still plaguing the general applicability of GNNs.