Reproduce “CLOSURE: Cross-Channel Closure for Multivariate Time Series Anomaly Detection” 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
Fan Xu, Hao Wu, Zhi Song, Hanbo Huang, Wenzhuo Ma, Nan Wang, Wei Gong, Xibin Zhao
CLOSURE detects anomalies in multivariate time series using two separately trained paths: one reconstructs each window, while the other predicts each channel from its peers. Calibrated scores are combined. On SMD, the full model reaches affiliation F1 of 88.13, versus 78.10 for reconstruction alone and 78.04 for D3R, across five-run benchmark results; evidence is limited to evaluated datasets.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.