Impute With Confidence: A Framework for Uncertainty Aware Multivariate Time Series Imputation

TL;DR

This paper introduces a framework for uncertainty-aware multivariate time series imputation, enabling selective imputation based on confidence levels to improve accuracy and downstream task performance.

Contribution

It presents a novel framework that quantifies and utilizes uncertainty in time series imputation, addressing limitations of existing methods that ignore model confidence.

Findings

Selective imputation reduces errors in EHR datasets.

Incorporating uncertainty improves mortality prediction accuracy.

Framework enhances downstream clinical decision-making.

Abstract

Time series data with missing values is common across many domains. Healthcare presents special challenges due to prolonged periods of sensor disconnection. In such cases, having a confidence measure for imputed values is critical. Most existing methods either overlook model uncertainty or lack mechanisms to estimate it. To address this gap, we introduce a general framework that quantifies and leverages uncertainty for selective imputation. By focusing on values the model is most confident in, highly unreliable imputations are avoided. Our experiments on multiple EHR datasets, covering diverse types of missingness, demonstrate that selectively imputing less-uncertain values not only reduces imputation errors but also improves downstream tasks. Specifically, we show performance gains in a 24-hour mortality prediction task, underscoring the practical benefit of incorporating uncertainty into time series imputation.