Get Started
Home
Topics
Search
Library
Research questionHow can representation-learning methods for dynamic heterogeneous graphs be compared across temporal granularities?Dynamic heterogeneous graphs combine changing structure with multiple entity and interaction types. Methods also make different assumptions about how time is represented, making their learned representations difficult to compare consistently.
AI
Evaluation & Benchmarks
Machine Learning
Research Paper
Latest papersRecent research connected to this question, newest first.Dynamic Heterogeneous Graph Representation Learning: A SurveyThe source surveys embedding-based, graph-neural-network, and Transformer-based approaches for dynamic heterogeneous graph representation learning. It organizes methods by algorithmic design and temporal granularity and discusses applications, datasets, and benchmarks, but does not provide evidence for a single universally preferred method.research paper · Sep 4, 2026
Related questions
How can learned models approximate trajectory similarity across multiple granularities without losing relative similarity rankings?How can reservoir-based graph neural networks aggregate multi-hop information without over-smoothing?How can learned graph simulators preserve contact history for accurate long-horizon granular-dynamics rollouts?How can graph neural networks capture long-range interactions without oversmoothing or oversquashing on diverse large graphs?