Random Dot Product Graph

Random Dot Product Graphs (RDPGs) model networks by representing nodes as vectors in a low-dimensional space, where the probability of an edge between two nodes is determined by the dot product of their corresponding vectors. Current research focuses on extending RDPGs to handle diverse network types (e.g., signed, weighted, directed, multiplex) and developing efficient algorithms for embedding and inference, including gradient-based methods and neural network approaches like GraphMoE. These advancements improve the accuracy and scalability of RDPG-based analyses, enabling applications in diverse fields such as brain network analysis and online change point detection in streaming graph data.