Online Learning Approach for Survival Analysis

Camila Fernandez (1) , Pierre Gaillard (2) , Joseph de Vilmarest (3) , Olivier Wintenberger (1)

1 LPSM (UMR_8001) - Laboratoire de Probabilités, Statistique et Modélisation
2 Thoth - Apprentissage de modèles à partir de données massives
3 Viking Conseil [Paris]

We introduce an online mathematical framework for survival analysis, allowing real time adaptation to dynamic environments and censored data. This framework enables the estimation of event time distributions through an optimal second order online convex optimization algorithm—Online Newton Step (ONS). This approach, previously unexplored, presents substantial advantages, including explicit algorithms with non-asymptotic convergence guarantees. Moreover, we analyze the selection of ONS hyperparameters, which depends on the exp-concavity property and has a significant influence on the regret bound. We propose a stochastic approach that guarantees logarithmic stochastic regret for ONS. Additionally, we introduce an adaptive aggregation method that ensures robustness in hyperparameter selection while maintaining fast regret bounds. The findings of this paper can extend beyond the survival analysis field, and are relevant for any case characterized by poor exp-concavity and unstable ONS. Finally, these assertions are illustrated by simulation experiments.

Mots clés: Online learning; survival analysis; regret bounds; convex optimization; stochastic risk

Domaines: Analyse de données, Statistiques et Probabilités [physics.data-an]

Camila Fernandez, Pierre Gaillard, Joseph de Vilmarest, Olivier Wintenberger. Online Learning Approach for Survival Analysis. 2024. ⟨hal-04418099⟩