PAST: A Primary-Auxiliary Spatio-Temporal Network for Traffic Time Series Imputation

TL;DR

This paper introduces PAST, a novel spatio-temporal network that effectively imputes missing traffic data by modeling both primary internal patterns and auxiliary external influences, outperforming existing methods.

Contribution

The paper proposes PAST, a new model combining graph-based primary pattern learning and external feature integration for improved traffic data imputation.

Findings

Outperforms seven state-of-the-art baselines in RMSE and MAE.

Effectively handles diverse missing data conditions.

Demonstrates robustness across three datasets with 27 missing data scenarios.

Abstract

Traffic time series imputation is crucial for the safety and reliability of intelligent transportation systems, while diverse types of missing data, including random, fiber, and block missing make the imputation task challenging. Existing models often focus on disentangling and separately modeling spatial and temporal patterns based on relationships between data points. However, these approaches struggle to adapt to the random missing positions, and fail to learn long-term and large-scale dependencies, which are essential in extensive missing conditions. In this paper, patterns are categorized into two types to handle various missing data conditions: primary patterns, which originate from internal relationships between data points, and auxiliary patterns, influenced by external factors like timestamps and node attributes. Accordingly, we propose the Primary-Auxiliary Spatio-Temporal network (PAST). It comprises a graph-integrated module (GIM) and a cross-gated module (CGM). GIM captures primary patterns via dynamic graphs with interval-aware dropout and multi-order convolutions, and CGM extracts auxiliary patterns through bidirectional gating on embedded external features. The two modules interact via shared hidden vectors and are trained under an ensemble self-supervised framework. Experiments on three datasets under 27 missing data conditions demonstrate that the imputation accuracy of PAST outperforms seven state-of-the-art baselines by up to 26.2% in RMSE and 31.6% in MAE.