English

Functional Decomposition and Shapley Interactions for Interpreting Survival Models

Machine Learning 2026-02-19 v1 Machine Learning

Abstract

Hazard and survival functions are natural, interpretable targets in time-to-event prediction, but their inherent non-additivity fundamentally limits standard additive explanation methods. We introduce Survival Functional Decomposition (SurvFD), a principled approach for analyzing feature interactions in machine learning survival models. By decomposing higher-order effects into time-dependent and time-independent components, SurvFD offers a previously unrecognized perspective on survival explanations, explicitly characterizing when and why additive explanations fail. Building on this theoretical decomposition, we propose SurvSHAP-IQ, which extends Shapley interactions to time-indexed functions, providing a practical estimator for higher-order, time-dependent interactions. Together, SurvFD and SurvSHAP-IQ establish an interaction- and time-aware interpretability approach for survival modeling, with broad applicability across time-to-event prediction tasks.

Keywords

Cite

@article{arxiv.2602.16505,
  title  = {Functional Decomposition and Shapley Interactions for Interpreting Survival Models},
  author = {Sophie Hanna Langbein and Hubert Baniecki and Fabian Fumagalli and Niklas Koenen and Marvin N. Wright and Julia Herbinger},
  journal= {arXiv preprint arXiv:2602.16505},
  year   = {2026}
}
R2 v1 2026-07-01T10:41:24.706Z