Estimating Missing Data in Temporal Data Streams Using Multi-directional Recurrent Neural Networks

23 Nov 2017Jinsung YoonWilliam R. ZameMihaela van der Schaar

Missing data is a ubiquitous problem. It is especially challenging in medical settings because many streams of measurements are collected at different - and often irregular - times... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
Multivariate Time Series Imputation Beijing Air Quality M-RNN MAE (PM2.5) 14.24 # 3
Multivariate Time Series Imputation PhysioNet Challenge 2012 M-RNN MAE (10% of data as GT) 0.451 # 3
Multivariate Time Series Imputation UCI localization data M-RNN MAE (10% missing) 0.248 # 2