Reproduce “A Study of Joint Graph Inference and Forecasting” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Daniel Zügner, François-Xavier Aubet, Victor Garcia Satorras, Tim Januschowski, Stephan Günnemann, Jan Gasthaus
The authors compare four models that learn graphs while forecasting multiple time series, testing how graph inputs affect prediction and whether learned edges match known relationships. On METR-LA, MTGNN’s horizon-12 MAE is 3.89 with the ground-truth graph, a 9.85% reduction versus vanilla, while GTS’s is 3.74, a 3.27% increase. These architecture-specific results do not establish a general benefit.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.