Adaptive Experimentation for Censored Survival Outcomes

TL;DR

This paper introduces a novel adaptive experimentation framework for censored survival data, optimizing causal effect estimation with theoretical guarantees and practical efficiency improvements.

Contribution

It develops a semiparametric efficiency bound, derives an optimal allocation policy, and proposes the Adaptive Survival Estimator (ASE) for sequential learning in censored survival trials.

Findings

ASE achieves consistent efficiency gains over baseline methods.

The framework provides asymptotic normality guarantees.

It accommodates machine learning models for nuisance estimation.

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.