Survival Function Matching for Calibrated Time-to-Event Predictions

Abstract

Models for predicting the time of a future event are crucial for risk assessment, across a diverse range of applications. Existing time-to-event (survival) models have focused primarily on preserving pairwise ordering of estimated event times, or relative risk. Model calibration is relatively under explored, despite its critical importance in time-to-event applications. We present a survival function estimator for probabilistic predictions in time-to-event models, based on a neural network model for draws from the distribution of event times, without explicit assumptions on the form of the distribution. This is done like in adversarial learning, but we achieve learning without a discriminator or adversarial objective. The proposed estimator can be used in practice as a means of estimating and comparing conditional survival distributions, while accounting for the predictive uncertainty of probabilistic models. Extensive experiments show that the proposed model outperforms existing approaches, trained both with and without adversarial learning, in terms of both calibration and concentration of time-to-event distributions.

References in corpus (12)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Auto-Encoding Variational Bayes
• Scalable and accurate deep learning for electronic health records
• Random survival forests
• DeepSurv: Personalized Treatment Recommender System Using A Cox Proportional Hazards Deep Neural Network
• On Calibration of Modern Neural Networks
• Accurate Uncertainties for Deep Learning Using Calibrated Regression
• Regularization for Cox's proportional hazards model with NP-dimensionality
• Deep Neural Networks for Survival Analysis Based on a Multi-Task Framework
• Gaussian Processes for Survival Analysis
• Countdown Regression: Sharp and Calibrated Survival Predictions
• Nonparametric Bayesian Lomax delegate racing for survival analysis with competing risks

Cited by in corpus (4)

• AdvMIL: Adversarial Multiple Instance Learning for the Survival Analysis on Whole-Slide Images
• Counterfactual Phenotyping with Censored Time-to-Events
• Enabling Counterfactual Survival Analysis with Balanced Representations
• Real-time Adaptation for Condition Monitoring Signal Prediction using Label-aware Neural Processes