Low-Rank Autoregressive Tensor Completion for Spatiotemporal Traffic Data Imputation (2021-04-30T00:00:00.000000Z)

TL;DR

A low-rank autoregressive tensor completion (LATC) framework is proposed by introducing temporal variation as a new regularization term into the completion of a third-order tensor structure to better capture the global consistency of traffic data, such as the inherent seasonality and day-to-day similarity.

Abstract

Spatiotemporal traffic time series (e.g., traffic volume/speed) collected from sensing systems are often incomplete with considerable corruption and large amounts of missing values, preventing users from harnessing the full power of the data. Missing data imputation has been a long-standing research topic and critical application for real-world intelligent transportation systems. A widely applied imputation method is low-rank matrix/tensor completion; however, the low-rank assumption only preserves the global structure while ignores the strong local consistency in spatiotemporal data. In this paper, we propose a low-rank autoregressive tensor completion (LATC) framework by introducing temporal variation as a new regularization term into the completion of a third-order (sensor $\times $ time of day $\times $ day) tensor. The third-order tensor structure allows us to better capture the global consistency of traffic data, such as the inherent seasonality and day-to-day similarity. To achieve local consistency, we design the temporal variation by imposing an autoregressive model for each time series with coefficients as learnable parameters. Different from previous spatial and temporal regularization schemes, the minimization of temporal variation can better characterize temporal generative mechanisms beyond local smoothness, allowing us to deal with more challenging scenarios such as “blackout” missing. To solve the optimization problem in LATC, we introduce an alternating minimization scheme that estimates the low-rank tensor and autoregressive coefficients iteratively. We conduct extensive numerical experiments on several real-world traffic data sets, and our results demonstrate the effectiveness of LATC in diverse missing scenarios.

Low-Rank Autoregressive Tensor Completion for Spatiotemporal Traffic Data Imputation

TL;DR

Abstract

Authors

References20 items

A Nonconvex Low-Rank Tensor Completion Model for Spatiotemporal Traffic Data Imputation

Tensor Completion for Weakly-dependent Data on Graph for Metro Passenger Flow Prediction

Bayesian Temporal Factorization for Multidimensional Time Series Prediction

Think Globally, Act Locally: A Deep Neural Network Approach to High-Dimensional Time Series Forecasting

A Bayesian tensor decomposition approach for spatiotemporal traffic data imputation

Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networks

Temporal Regularized Matrix Factorization for High-dimensional Time Series Prediction

Short-Term Traffic Prediction Based on Dynamic Tensor Completion

Collaborative Filtering with Graph Information: Consistency and Scalable Methods

Trend Modeling for Traffic Time Series Analysis: An Integrated Study

Smooth PARAFAC Decomposition for Tensor Completion

Fast Multivariate Spatio-temporal Analysis via Low Rank Tensor Learning

Fast and Accurate Matrix Completion via Truncated Nuclear Norm Regularization

Reduced rank regression via adaptive nuclear norm penalization.

Penalty decomposition methods for rank minimization

Tensor Completion for Estimating Missing Values in Visual Data

Tensor Decompositions and Applications

DynaMMo: mining and summarization of coevolving sequences with missing values

Temporal Collaborative Filtering with Bayesian Probabilistic Tensor Factorization

We develop an alternating learning algorithm to update tensor and coefﬁcient matrix separately

Field of Study

Journal Information

Name

Page

Volume