Adaptive Experimentation for Censored Survival Outcomes

2026-05-18

AI Summary

This work addresses the challenge of efficiently estimating causal effects under right-censored survival data in adaptive experimental settings. The authors propose the Adaptive Survival Estimator (ASE), which first derives the semiparametric efficiency bound for the average survival effect curve and obtains a closed-form, efficiency-optimal allocation strategy. Building on this, they extend the classical Neyman allocation to survival analysis by explicitly incorporating the censoring mechanism and event dynamics, yielding a novel uncertainty-aware sequential allocation scheme that accommodates any machine learning model for nuisance parameter estimation. Theoretical analysis establishes the asymptotic normality of ASE, and numerical experiments demonstrate its consistent efficiency gains over both uniform randomization and baseline methods that ignore censoring.

Abstract

Adaptive experimentation enables efficient estimation of causal effects, but existing methods are not designed for survival data with censoring, where event times are only partially observed (e.g., overall survival in cancer trials but with dropout). In this paper, we develop a novel framework for adaptive experimentation to estimate causal effects under right censoring. For this, we derive the semiparametric efficiency bound for the average survival effect curve as a function of the treatment allocation policy and thereby obtain a closed-form efficiency-optimal allocation policy. The policy generalizes classical Neyman allocation to survival settings by prioritizing patient strata where both event and censoring dynamics induce high uncertainty. Building on this, we propose the Adaptive Survival Estimator (ASE), an adaptive framework that learns the allocation policy and estimates the average survival effect curve sequentially. Our framework has three main benefits: (i) it accommodates arbitrary machine learning models for nuisance estimation; (ii) it is guided by a closed-form efficiency-optimal allocation policy; and (iii) it admits strong theoretical guarantees, including asymptotic normality via a martingale central limit theorem. We demonstrate our framework across various numerical experiments to show consistent efficiency gains over uniform randomization and censoring-agnostic baselines.

Problem

• adaptive experimentation
• censored survival outcomes
• causal effect estimation
• right censoring
• survival analysis

Innovation

• adaptive experimentation
• censored survival data
• efficiency-optimal allocation
• semiparametric efficiency
• causal inference