Streaming data recovery via Bayesian tensor train decomposition

arXiv:2302.12148v2 Announce Type: replace
Abstract: In this paper, we study a Bayesian tensor train (TT) decomposition method to recover streaming data by approximating the latent structure in high-order streaming data. Drawing on the streaming variational Bayes method, we introduce the TT format into Bayesian tensor decomposition methods for streaming data, and formulate posteriors of TT cores. Thanks to the Bayesian framework of the TT format, the proposed algorithm (SPTT) excels in recovering streaming data with high-order, incomplete, and noisy properties. …

abstract arxiv bayes bayesian cs.lg data data recovery format math.st paper recovery stat.ml stat.th streaming streaming data study tensor train type via