Spatiotemporal Bayesian Online Changepoint Detection with Model Selection
[edit]
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:27182727, 2018.
Abstract
Bayesian Online Changepoint Detection is extended to online model selection and nonstationary spatiotemporal processes. We propose spatially structured Vector Autoregressions (VARs) for modelling the process between changepoints (CPs) and give an upper bound on the approximation error of such models. The resulting algorithm performs prediction, model selection and CP detection online. Its time complexity is linear and its space complexity constant, and thus it is two orders of magnitudes faster than its closest competitor. In addition, it outperforms the state of the art for multivariate data.
Related Material


