Adversarial Time-to-Event Modeling
This work addresses a key problem in health data science for clinical practice by improving time-to-event modeling, though it appears incremental as it builds on existing adversarial learning methods.
The paper tackles the challenge of nonparametric estimation of event-time distributions in time-to-event analysis by introducing a deep-network-based approach using adversarial learning and a principled cost function for censored events. The model demonstrates significant performance gains over a parametric alternative on benchmark and real datasets.
Modern health data science applications leverage abundant molecular and electronic health data, providing opportunities for machine learning to build statistical models to support clinical practice. Time-to-event analysis, also called survival analysis, stands as one of the most representative examples of such statistical models. We present a deep-network-based approach that leverages adversarial learning to address a key challenge in modern time-to-event modeling: nonparametric estimation of event-time distributions. We also introduce a principled cost function to exploit information from censored events (events that occur subsequent to the observation window). Unlike most time-to-event models, we focus on the estimation of time-to-event distributions, rather than time ordering. We validate our model on both benchmark and real datasets, demonstrating that the proposed formulation yields significant performance gains relative to a parametric alternative, which we also propose.