Reproduce “A New Perspective on Time Series Anomaly Detection: Faster Patch-based Broad Learning System” 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
Pengyu Li, Zhijie Zhong, Tong Zhang, Zhiwen Yu, C.L. Philip Chen, Kaixiang Yang
CPatchBLS detects anomalies in multivariate time series using a shallow Broad Learning System (BLS) with non-overlapping temporal patches. It compares basic and randomly perturbed representations, then averages anomaly scores across patch sizes. Across five real-world datasets, it reports averages of 99.81% ROC-AUC, 98.25% ROC-PR, and 96.87% PA-F1; these results do not establish universal superiority.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.