Spatio-temporal Bayesian On-line Changepoint Detection with Model Selection

Jeremias Knoblauch, Theodoros Damoulas
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:2718-2727, 2018.

Abstract

Bayesian On-line Changepoint Detection is extended to on-line model selection and non-stationary spatio-temporal 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 on-line. 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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v80-knoblauch18a, title = {Spatio-temporal {B}ayesian On-line Changepoint Detection with Model Selection}, author = {Knoblauch, Jeremias and Damoulas, Theodoros}, booktitle = {Proceedings of the 35th International Conference on Machine Learning}, pages = {2718--2727}, year = {2018}, editor = {Dy, Jennifer and Krause, Andreas}, volume = {80}, series = {Proceedings of Machine Learning Research}, month = {10--15 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v80/knoblauch18a/knoblauch18a.pdf}, url = {https://proceedings.mlr.press/v80/knoblauch18a.html}, abstract = {Bayesian On-line Changepoint Detection is extended to on-line model selection and non-stationary spatio-temporal 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 on-line. 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.} }
Endnote
%0 Conference Paper %T Spatio-temporal Bayesian On-line Changepoint Detection with Model Selection %A Jeremias Knoblauch %A Theodoros Damoulas %B Proceedings of the 35th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2018 %E Jennifer Dy %E Andreas Krause %F pmlr-v80-knoblauch18a %I PMLR %P 2718--2727 %U https://proceedings.mlr.press/v80/knoblauch18a.html %V 80 %X Bayesian On-line Changepoint Detection is extended to on-line model selection and non-stationary spatio-temporal 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 on-line. 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.
APA
Knoblauch, J. & Damoulas, T.. (2018). Spatio-temporal Bayesian On-line Changepoint Detection with Model Selection. Proceedings of the 35th International Conference on Machine Learning, in Proceedings of Machine Learning Research 80:2718-2727 Available from https://proceedings.mlr.press/v80/knoblauch18a.html.

Related Material