🤖 AI Summary
To address the neglect of dynamic spatial dependencies in multivariate time series imputation, this paper proposes a Dynamic Spatio-Temporal Graph Neural Network (DST-GNN). The method models multivariate time series as time-varying graphs, where spatial topology is adaptively learned via multi-head attention mechanisms to capture temporal evolution of spatial relationships; recurrent message passing further enables joint spatio-temporal modeling. The key contribution is the first introduction of a spatial dynamic awareness mechanism that explicitly characterizes time-varying spatial dependencies, integrated within a unified framework for time-varying graph construction and spatio-temporal modeling. Extensive experiments on AQI, AQI-36, and PEMS-BAY datasets demonstrate significant improvements: MSE reductions of 9.51%, 9.40%, and 1.94%, respectively, outperforming state-of-the-art methods.
📝 Abstract
In various applications, the multivariate time series often suffers from missing data. This issue can significantly disrupt systems that rely on the data. Spatial and temporal dependencies can be leveraged to impute the missing samples. Existing imputation methods often ignore dynamic changes in spatial dependencies. We propose a Spatial Dynamic Aware Graph Recurrent Imputation Network (SDA-GRIN) which is capable of capturing dynamic changes in spatial dependencies.SDA-GRIN leverages a multi-head attention mechanism to adapt graph structures with time. SDA-GRIN models multivariate time series as a sequence of temporal graphs and uses a recurrent message-passing architecture for imputation. We evaluate SDA-GRIN on four real-world datasets: SDA-GRIN improves MSE by 9.51% for the AQI and 9.40% for AQI-36. On the PEMS-BAY dataset, it achieves a 1.94% improvement in MSE. Detailed ablation study demonstrates the effect of window sizes and missing data on the performance of the method. Project page:https://ameskandari.github.io/sda-grin/