On the calibration of survival models with competing risks

Julie Alberge, Tristan Haugomat, Gaël Varoquaux, Judith Abécassis
Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, PMLR 300:1837-1845, 2026.

Abstract

Survival analysis deals with modeling the time until an event occurs, and accurate probability estimates are crucial for decision-making, particularly in the competing-risks setting where multiple events are possible. While recent work has addressed calibration in standard survival analysis, the competing-risks setting remains under-explored as it is harder (the calibration applies to both probabilities across classes and time horizon). We show that existing calibration measures are not suited to the competing-risk setting and that recent models do not give well-behaved probabilities. To address this, we introduce a dedicated framework with two novel calibration measures that are minimized for oracle estimators (\emph{i.e.}, both measures are proper). We also introduce some methods to estimate, test, and correct the calibration. Our recalibration methods yield better probabilities while preserving discrimination.

Cite this Paper

BibTeX
@InProceedings{pmlr-v300-alberge26a, title = { On the calibration of survival models with competing risks }, author = {Alberge, Julie and Haugomat, Tristan and Varoquaux, Ga{\"e}l and Ab{\'e}cassis, Judith}, booktitle = {Proceedings of The 29th International Conference on Artificial Intelligence and Statistics}, pages = {1837--1845}, year = {2026}, editor = {Khan, Emtiyaz and Li, Yingzhen and Solin, Arno and Ramdas, Aaditya}, volume = {300}, series = {Proceedings of Machine Learning Research}, month = {02--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v300/main/assets/alberge26a/alberge26a.pdf}, url = {https://proceedings.mlr.press/v300/alberge26a.html}, abstract = { Survival analysis deals with modeling the time until an event occurs, and accurate probability estimates are crucial for decision-making, particularly in the competing-risks setting where multiple events are possible. While recent work has addressed calibration in standard survival analysis, the competing-risks setting remains under-explored as it is harder (the calibration applies to both probabilities across classes and time horizon). We show that existing calibration measures are not suited to the competing-risk setting and that recent models do not give well-behaved probabilities. To address this, we introduce a dedicated framework with two novel calibration measures that are minimized for oracle estimators (\emph{i.e.}, both measures are proper). We also introduce some methods to estimate, test, and correct the calibration. Our recalibration methods yield better probabilities while preserving discrimination. } }
Endnote
%0 Conference Paper %T On the calibration of survival models with competing risks %A Julie Alberge %A Tristan Haugomat %A Gaël Varoquaux %A Judith Abécassis %B Proceedings of The 29th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2026 %E Emtiyaz Khan %E Yingzhen Li %E Arno Solin %E Aaditya Ramdas %F pmlr-v300-alberge26a %I PMLR %P 1837--1845 %U https://proceedings.mlr.press/v300/alberge26a.html %V 300 %X Survival analysis deals with modeling the time until an event occurs, and accurate probability estimates are crucial for decision-making, particularly in the competing-risks setting where multiple events are possible. While recent work has addressed calibration in standard survival analysis, the competing-risks setting remains under-explored as it is harder (the calibration applies to both probabilities across classes and time horizon). We show that existing calibration measures are not suited to the competing-risk setting and that recent models do not give well-behaved probabilities. To address this, we introduce a dedicated framework with two novel calibration measures that are minimized for oracle estimators (\emph{i.e.}, both measures are proper). We also introduce some methods to estimate, test, and correct the calibration. Our recalibration methods yield better probabilities while preserving discrimination.
APA
Alberge, J., Haugomat, T., Varoquaux, G. & Abécassis, J.. (2026). On the calibration of survival models with competing risks . Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 300:1837-1845 Available from https://proceedings.mlr.press/v300/alberge26a.html.

Related Material