Compatible Transformer for Irregularly Sampled Multivariate Time Series
This addresses a practical issue for data collection systems in fields like healthcare or IoT, where irregular sampling is common, though it is an incremental advancement over transformer-based methods.
The paper tackles the problem of analyzing irregularly sampled multivariate time series, which arise from sensor failures and interventions, by proposing Compatible Transformer (CoFormer) to handle misalignment in temporal and variate dimensions, achieving significant and consistent performance improvements over existing methods on 3 real-world datasets.
To analyze multivariate time series, most previous methods assume regular subsampling of time series, where the interval between adjacent measurements and the number of samples remain unchanged. Practically, data collection systems could produce irregularly sampled time series due to sensor failures and interventions. However, existing methods designed for regularly sampled multivariate time series cannot directly handle irregularity owing to misalignment along both temporal and variate dimensions. To fill this gap, we propose Compatible Transformer (CoFormer), a transformer-based encoder to achieve comprehensive temporal-interaction feature learning for each individual sample in irregular multivariate time series. In CoFormer, we view each sample as a unique variate-time point and leverage intra-variate/inter-variate attentions to learn sample-wise temporal/interaction features based on intra-variate/inter-variate neighbors. With CoFormer as the core, we can analyze irregularly sampled multivariate time series for many downstream tasks, including classification and prediction. We conduct extensive experiments on 3 real-world datasets and validate that the proposed CoFormer significantly and consistently outperforms existing methods.