Temporal Parallelization of Bayesian Smoothers

• 成果类型: Article
• 署名作者: Sarkka, Simo; Garcia-Fernandez, Angel F.
• 署名单位: Aalto University; University of Liverpool
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2020.2976316
• 发表日期: 2021
• 页码: 299-306
• 关键词: Bayes methods; Smoothing methods; Mathematical model; computational modeling; Kalman filters; parallel algorithms; Bayesian smoothing; Kalman filtering and smoothing; parallel computing; parallel scan; prefix sums

摘要

This article presents algorithms for temporal parallelization of Bayesian smoothers. We define the elements and the operators to pose these problems as the solutions to all-prefix-sums operations for which efficient parallel scan-algorithms are available. We present the temporal parallelization of the general Bayesian filtering and smoothing equations, and specialize them to linear/Gaussian models. The advantage of the proposed algorithms is that they reduce the linear complexity of standard smoothing algorithms with respect to time to logarithmic.