A0415
Title: Topological analysis for detecting anomalies in time series
Authors: Clement Levrard - Université de Rennes (France) [presenting]
Abstract: A recent methodology is exposed based on the field of topological data analysis for detecting anomalies in multivariate time series, which aims to detect global changes in the dependency structure between channels. This approach is lean enough to handle large-scale datasets, and extensive numerical experiments back the intuition that it is more suitable for detecting global changes of correlation structures than existing methods. If time allows, some theoretical guarantees will also be presented for this method.
