Representation Learning for Dynamic Graphs: A Survey

TL;DR

This survey reviews recent advances in representation learning techniques for dynamic graphs, highlighting models, applications, datasets, and future research directions in evolving graph analysis.

Contribution

It categorizes dynamic graph models from an encoder-decoder perspective and analyzes their techniques and applications.

Findings

Categorization of dynamic graph models based on encoding and decoding techniques

Analysis of applications and datasets used in dynamic graph learning

Identification of future research directions in the field

Abstract

Graphs arise naturally in many real-world applications including social networks, recommender systems, ontologies, biology, and computational finance. Traditionally, machine learning models for graphs have been mostly designed for static graphs. However, many applications involve evolving graphs. This introduces important challenges for learning and inference since nodes, attributes, and edges change over time. In this survey, we review the recent advances in representation learning for dynamic graphs, including dynamic knowledge graphs. We describe existing models from an encoder-decoder perspective, categorize these encoders and decoders based on the techniques they employ, and analyze the approaches in each category. We also review several prominent applications and widely used datasets and highlight directions for future research.