Copula-Based Deep Survival Models for Dependent Censoring
This addresses bias in survival analysis for fields like healthcare, where censoring dependence is common but often ignored, though it is incremental as it builds on existing non-linear survival models.
The paper tackles the problem of survival prediction under dependent censoring, where standard methods assume conditional independence between event and censoring times, by introducing a parametric model that relaxes this assumption and significantly improves survival distribution estimates on synthetic and semi-synthetic data.
A survival dataset describes a set of instances (e.g. patients) and provides, for each, either the time until an event (e.g. death), or the censoring time (e.g. when lost to follow-up - which is a lower bound on the time until the event). We consider the challenge of survival prediction: learning, from such data, a predictive model that can produce an individual survival distribution for a novel instance. Many contemporary methods of survival prediction implicitly assume that the event and censoring distributions are independent conditional on the instance's covariates - a strong assumption that is difficult to verify (as we observe only one outcome for each instance) and which can induce significant bias when it does not hold. This paper presents a parametric model of survival that extends modern non-linear survival analysis by relaxing the assumption of conditional independence. On synthetic and semi-synthetic data, our approach significantly improves estimates of survival distributions compared to the standard that assumes conditional independence in the data.