The two-stage non-parametric algorithm for change point detection in multivariate time series proceeds by: (1) learning parts of the causal structure through constraint-based discovery methods, and (2) employing conditional relative Pearson divergence estimation to identify the change points.

Authors

• Person: Samuel Tesfazgi, Leonhard Sprandl, Sandra Hirche Organization: AISTATS
  Track: Poster Session 3 - aistats 2026

Sources

• Track: Poster Session 3 - aistats 2026 virtual.aistats.org Samuel Tesfazgi, Leonhard Sprandl, Sandra Hirche · AISTATS via serper

Referenced by nodes (1)

• causal structure concept