Continuous Time Evidential Distributions for Irregular Time Series

Taylor W. Killian, Haoran Zhang, Thomas Hartvigsen, Ava P Amini

Keywords: Evidential Deep Learning, Irregular Time Series, Uncertainty Quantification, Continuous Time Models, Healthcare

TL;DR: We develop a continuous time, evidential distribution for processing irregular time series which enables temporally correlated measures of uncertainty over intervals of missingness.

Abstract: Prevalent in many real-world settings such as healthcare, irregular time series are challenging to formulate predictions from. It is difficult to infer the value of a feature at any given time when observations are sporadic, as it could take on a range of values depending on when it was last observed. To characterize this uncertainty we present EDICT, a strategy that learns an evidential distribution over irregular time series in continuous time. This distribution enables well-calibrated and flexible inference of partially observed features at any time of interest, while expanding uncertainty temporally for sparse, irregular observations. We demonstrate that EDICT attains competitive performance on challenging time series classification tasks and enabling uncertainty-guided inference when encountering noisy data.

Submission Number: 37