Recurrent Neural Networks for Multivariate Time Series with Missing Values

TL;DR

This paper introduces GRU-D, a novel deep learning model that leverages missing data patterns in multivariate time series to improve prediction accuracy, especially in healthcare datasets with informative missingness.

Contribution

The paper proposes GRU-D, an innovative recurrent neural network that incorporates missing data patterns into its architecture to enhance time series prediction performance.

Findings

Achieves state-of-the-art results on clinical datasets

Effectively utilizes missing patterns for better predictions

Provides insights into missing data utilization

Abstract

Multivariate time series data in practical applications, such as health care, geoscience, and biology, are characterized by a variety of missing values. In time series prediction and other related tasks, it has been noted that missing values and their missing patterns are often correlated with the target labels, a.k.a., informative missingness. There is very limited work on exploiting the missing patterns for effective imputation and improving prediction performance. In this paper, we develop novel deep learning models, namely GRU-D, as one of the early attempts. GRU-D is based on Gated Recurrent Unit (GRU), a state-of-the-art recurrent neural network. It takes two representations of missing patterns, i.e., masking and time interval, and effectively incorporates them into a deep model architecture so that it not only captures the long-term temporal dependencies in time series, but also utilizes the missing patterns to achieve better prediction results. Experiments of time series classification tasks on real-world clinical datasets (MIMIC-III, PhysioNet) and synthetic datasets demonstrate that our models achieve state-of-the-art performance and provides useful insights for better understanding and utilization of missing values in time series analysis.