A1096
Title: Learning robust treatment rules for censored data
Authors: Yifan Cui - Zhejiang University (China)
Junyi Liu - Tsinghua University (China)
Tao Shen - National University of Singapore (China) [presenting]
Zhengling Qi - The George Washington University (United States)
Xi Chen - New York University (United States)
Abstract: There is a fast-growing literature on estimating optimal treatment rules directly by maximizing the expected outcome. In biomedical studies and operations applications, censored survival outcome is frequently observed, in which case the restricted mean survival time and survival probability are of great interest. Two robust criteria are proposed for learning optimal treatment rules with censored survival outcomes; the former one targets an optimal treatment rule maximizing the restricted mean survival time, where the restriction is specified by a given quantile such as median; the latter one targets at an optimal treatment rule maximizing buffered survival probabilities, where the predetermined threshold is adjusted to account the restricted mean survival time. Theoretical justifications are provided for the proposed optimal treatment rules, and a sampling-based difference-of-convex algorithm is developed for learning them. The proposed method is also demonstrated using AIDS clinical trial data.
