Survival Models: Proper Scoring Rule and Stochastic Optimization with Competing Risks
This addresses scalability and efficiency issues in survival analysis for domains like healthcare, offering a novel method for competing risks prediction.
The paper tackles the problem of predicting time-to-event outcomes with competing risks in right-censored data by designing a proper scoring rule that enables stochastic optimization, leading to SurvivalBoost, which outperforms 12 state-of-the-art models on 4 real-life datasets with faster computation times.
When dealing with right-censored data, where some outcomes are missing due to a limited observation period, survival analysis -- known as time-to-event analysis -- focuses on predicting the time until an event of interest occurs. Multiple classes of outcomes lead to a classification variant: predicting the most likely event, a less explored area known as competing risks. Classic competing risks models couple architecture and loss, limiting scalability.To address these issues, we design a strictly proper censoring-adjusted separable scoring rule, allowing optimization on a subset of the data as each observation is evaluated independently. The loss estimates outcome probabilities and enables stochastic optimization for competing risks, which we use for efficient gradient boosting trees. SurvivalBoost not only outperforms 12 state-of-the-art models across several metrics on 4 real-life datasets, both in competing risks and survival settings, but also provides great calibration, the ability to predict across any time horizon, and computation times faster than existing methods.