Adversarial Time-to-Event Modeling

Paidamoyo Chapfuwa, Chenyang Tao, Chunyuan Li, Courtney Page, Benjamin Goldstein, Lawrence Carin Duke, Ricardo Henao
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:735-744, 2018.

Abstract

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v80-chapfuwa18a, title = {Adversarial Time-to-Event Modeling}, author = {Chapfuwa, Paidamoyo and Tao, Chenyang and Li, Chunyuan and Page, Courtney and Goldstein, Benjamin and Duke, Lawrence Carin and Henao, Ricardo}, booktitle = {Proceedings of the 35th International Conference on Machine Learning}, pages = {735--744}, year = {2018}, editor = {Dy, Jennifer and Krause, Andreas}, volume = {80}, series = {Proceedings of Machine Learning Research}, month = {10--15 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v80/chapfuwa18a/chapfuwa18a.pdf}, url = {https://proceedings.mlr.press/v80/chapfuwa18a.html}, abstract = {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.} }
Endnote
%0 Conference Paper %T Adversarial Time-to-Event Modeling %A Paidamoyo Chapfuwa %A Chenyang Tao %A Chunyuan Li %A Courtney Page %A Benjamin Goldstein %A Lawrence Carin Duke %A Ricardo Henao %B Proceedings of the 35th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2018 %E Jennifer Dy %E Andreas Krause %F pmlr-v80-chapfuwa18a %I PMLR %P 735--744 %U https://proceedings.mlr.press/v80/chapfuwa18a.html %V 80 %X 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.
APA
Chapfuwa, P., Tao, C., Li, C., Page, C., Goldstein, B., Duke, L.C. & Henao, R.. (2018). Adversarial Time-to-Event Modeling. Proceedings of the 35th International Conference on Machine Learning, in Proceedings of Machine Learning Research 80:735-744 Available from https://proceedings.mlr.press/v80/chapfuwa18a.html.

Related Material