Forecasting Irregularly Sampled Time Series using Graphs
This addresses a crucial problem in domains like healthcare and climate sciences by offering a faster and more accurate alternative to ODE-based methods, though it appears incremental as it builds on graph neural networks for a known bottleneck.
The paper tackles forecasting irregularly sampled time series with missing values by proposing GraFITi, a model that converts time series to a sparsity structure graph and uses graph neural networks for edge weight prediction, resulting in up to 17% accuracy improvement and 5x runtime reduction compared to state-of-the-art models.
Forecasting irregularly sampled time series with missing values is a crucial task for numerous real-world applications such as healthcare, astronomy, and climate sciences. State-of-the-art approaches to this problem rely on Ordinary Differential Equations (ODEs) which are known to be slow and often require additional features to handle missing values. To address this issue, we propose a novel model using Graphs for Forecasting Irregularly Sampled Time Series with missing values which we call GraFITi. GraFITi first converts the time series to a Sparsity Structure Graph which is a sparse bipartite graph, and then reformulates the forecasting problem as the edge weight prediction task in the graph. It uses the power of Graph Neural Networks to learn the graph and predict the target edge weights. GraFITi has been tested on 3 real-world and 1 synthetic irregularly sampled time series dataset with missing values and compared with various state-of-the-art models. The experimental results demonstrate that GraFITi improves the forecasting accuracy by up to 17% and reduces the run time up to 5 times compared to the state-of-the-art forecasting models.