CDSA: Cross-Dimensional Self-Attention for Multivariate, Geo-tagged Time Series Imputation

TL;DR

This paper introduces CDSA, a novel self-attention-based method for imputing missing values in multivariate, geo-tagged time series data, outperforming existing RNN-based approaches by modeling relationships across time, location, and sensor measurements.

Contribution

We propose Cross-Dimensional Self-Attention (CDSA), the first self-attention model for multivariate, geo-tagged time series, capturing cross-dimensional dependencies efficiently and effectively.

Findings

Outperforms state-of-the-art imputation methods on multiple datasets

Demonstrates superior accuracy in time series forecasting tasks

Validates effectiveness through extensive experiments and systematic analysis

Abstract

Many real-world applications involve multivariate, geo-tagged time series data: at each location, multiple sensors record corresponding measurements. For example, air quality monitoring system records PM2.5, CO, etc. The resulting time-series data often has missing values due to device outages or communication errors. In order to impute the missing values, state-of-the-art methods are built on Recurrent Neural Networks (RNN), which process each time stamp sequentially, prohibiting the direct modeling of the relationship between distant time stamps. Recently, the self-attention mechanism has been proposed for sequence modeling tasks such as machine translation, significantly outperforming RNN because the relationship between each two time stamps can be modeled explicitly. In this paper, we are the first to adapt the self-attention mechanism for multivariate, geo-tagged time series data. In order to jointly capture the self-attention across multiple dimensions, including time, location and the sensor measurements, while maintain low computational complexity, we propose a novel approach called Cross-Dimensional Self-Attention (CDSA) to process each dimension sequentially, yet in an order-independent manner. Our extensive experiments on four real-world datasets, including three standard benchmarks and our newly collected NYC-traffic dataset, demonstrate that our approach outperforms the state-of-the-art imputation and forecasting methods. A detailed systematic analysis confirms the effectiveness of our design choices.