Off-Policy Evaluation and Learning for Survival Outcomes under Censoring
This addresses a critical gap in data-driven decision-making for high-stakes applications like healthcare and customer retention, though it is an incremental improvement by adapting existing techniques to handle censoring.
The paper tackles the problem of off-policy evaluation and learning for survival outcomes under censoring, where existing estimators fail due to unobserved survival times, and proposes novel IPCW-based estimators that are unbiased and double robust, showing effectiveness in simulations and real-world data.
Optimizing survival outcomes, such as patient survival or customer retention, is a critical objective in data-driven decision-making. Off-Policy Evaluation~(OPE) provides a powerful framework for assessing such decision-making policies using logged data alone, without the need for costly or risky online experiments in high-stakes applications. However, typical estimators are not designed to handle right-censored survival outcomes, as they ignore unobserved survival times beyond the censoring time, leading to systematic underestimation of the true policy performance. To address this issue, we propose a novel framework for OPE and Off-Policy Learning~(OPL) tailored for survival outcomes under censoring. Specifically, we introduce IPCW-IPS and IPCW-DR, which employ the Inverse Probability of Censoring Weighting technique to explicitly deal with censoring bias. We theoretically establish that our estimators are unbiased and that IPCW-DR achieves double robustness, ensuring consistency if either the propensity score or the outcome model is correct. Furthermore, we extend this framework to constrained OPL to optimize policy value under budget constraints. We demonstrate the effectiveness of our proposed methods through simulation studies and illustrate their practical impacts using public real-world data for both evaluation and learning tasks.