TY - JOUR
T1 - Recurrent Neural Networks for Multivariate Time Series with Missing Values
AU - Che, Zhengping
AU - Purushotham, Sanjay
AU - Cho, Kyunghyun
AU - Sontag, David
AU - Liu, Yan
N1 - Publisher Copyright:
© 2018 The Author(s).
PY - 2018/12/1
Y1 - 2018/12/1
N2 - 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 provide useful insights for better understanding and utilization of missing values in time series analysis.
AB - 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 provide useful insights for better understanding and utilization of missing values in time series analysis.
UR - http://www.scopus.com/inward/record.url?scp=85045746406&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85045746406&partnerID=8YFLogxK
U2 - 10.1038/s41598-018-24271-9
DO - 10.1038/s41598-018-24271-9
M3 - Article
C2 - 29666385
AN - SCOPUS:85045746406
SN - 2045-2322
VL - 8
JO - Scientific reports
JF - Scientific reports
IS - 1
M1 - 6085
ER -