Reproduce “GT-PSSM: Unified Probabilistic Framework for Stochastic Dynamics Modeling and Dependency Learning in 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
Wonmo Koo, Jaeyeong Lee, Taeseong Yoon, Heeyoung Kim
GT-PSSM detects anomalies in multivariate time series by modeling expected readings and their uncertainty in a probabilistic state-space model. It uses temporal self-attention and graph convolution over learned variable relationships. On WADI, its unadjusted F1 was 0.7343 versus 0.6749 for SensitiveHUE, averaged over 10 runs. Across four benchmarks, it led on three and ranked second on SWaT.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.