Reproduce “SIMBAD: Spatio-Temporal Traffic Forecasting Robust to Aperiodicity” 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 Y. Lee, Seungwon Shin, Seunghoon Han, Sungsu Lim, Susik Yoon
SIMBAD forecasts traffic by treating historical patterns as clues, not fixed templates. It compares recent conditions with daily and weekly references, discounts unreliable histories, and adapts how connected sensors share information. On PEMS07, test MAE is 17.86 versus STIDGCN’s 19.20, averaged across predicted horizons; on METR-LA, SIMBAD trails LHMformer at 60 minutes, so results vary by dataset.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.