Spectral Subsampling MCMC for Stationary Time Series

Robert Salomone, Matias Quiroz, Robert Kohn, Mattias Villani, Minh-Ngoc Tran
Proceedings of the 37th International Conference on Machine Learning, PMLR 119:8449-8458, 2020.

Abstract

Bayesian inference using Markov Chain Monte Carlo (MCMC) on large datasets has developed rapidly in recent years. However, the underlying methods are generally limited to relatively simple settings where the data have specific forms of independence. We propose a novel technique for speeding up MCMC for time series data by efficient data subsampling in the frequency domain. For several challenging time series models, we demonstrate a speedup of up to two orders of magnitude while incurring negligible bias compared to MCMC on the full dataset. We also propose alternative control variates for variance reduction based on data grouping and coreset constructions.

Cite this Paper


BibTeX
@InProceedings{pmlr-v119-salomone20a, title = {Spectral Subsampling {MCMC} for Stationary Time Series}, author = {Salomone, Robert and Quiroz, Matias and Kohn, Robert and Villani, Mattias and Tran, Minh-Ngoc}, booktitle = {Proceedings of the 37th International Conference on Machine Learning}, pages = {8449--8458}, year = {2020}, editor = {III, Hal Daumé and Singh, Aarti}, volume = {119}, series = {Proceedings of Machine Learning Research}, month = {13--18 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v119/salomone20a/salomone20a.pdf}, url = {https://proceedings.mlr.press/v119/salomone20a.html}, abstract = {Bayesian inference using Markov Chain Monte Carlo (MCMC) on large datasets has developed rapidly in recent years. However, the underlying methods are generally limited to relatively simple settings where the data have specific forms of independence. We propose a novel technique for speeding up MCMC for time series data by efficient data subsampling in the frequency domain. For several challenging time series models, we demonstrate a speedup of up to two orders of magnitude while incurring negligible bias compared to MCMC on the full dataset. We also propose alternative control variates for variance reduction based on data grouping and coreset constructions.} }
Endnote
%0 Conference Paper %T Spectral Subsampling MCMC for Stationary Time Series %A Robert Salomone %A Matias Quiroz %A Robert Kohn %A Mattias Villani %A Minh-Ngoc Tran %B Proceedings of the 37th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2020 %E Hal Daumé III %E Aarti Singh %F pmlr-v119-salomone20a %I PMLR %P 8449--8458 %U https://proceedings.mlr.press/v119/salomone20a.html %V 119 %X Bayesian inference using Markov Chain Monte Carlo (MCMC) on large datasets has developed rapidly in recent years. However, the underlying methods are generally limited to relatively simple settings where the data have specific forms of independence. We propose a novel technique for speeding up MCMC for time series data by efficient data subsampling in the frequency domain. For several challenging time series models, we demonstrate a speedup of up to two orders of magnitude while incurring negligible bias compared to MCMC on the full dataset. We also propose alternative control variates for variance reduction based on data grouping and coreset constructions.
APA
Salomone, R., Quiroz, M., Kohn, R., Villani, M. & Tran, M.. (2020). Spectral Subsampling MCMC for Stationary Time Series. Proceedings of the 37th International Conference on Machine Learning, in Proceedings of Machine Learning Research 119:8449-8458 Available from https://proceedings.mlr.press/v119/salomone20a.html.

Related Material