Graph Prediction

Graph prediction focuses on leveraging machine learning to forecast or generate graphs, encompassing tasks like predicting future network structures, inferring missing edges, or generating graphs from other data types. Current research emphasizes the development of scalable and accurate models, including graph neural networks (GNNs) and optimal transport-based methods, often enhanced by techniques like self-supervised learning and incorporating edge features. These advancements are driving progress in diverse fields, from cybersecurity threat prediction and drug discovery to traffic flow analysis and the design of more efficient neural networks.