一种用于估计最优动态治疗方案的惩罚共享参数算法
A Penalized Shared-parameter Algorithm for Estimating Optimal Dynamic Treatment Regimens
AI总结:
针对 Q-shared 算法因 Q-learning 中线性模型导致的不收敛问题,提出惩罚 Q-shared 算法,明确失效条件,并通过真实应用与模拟验证其收敛性和更优性能。
摘要:
动态治疗方案(DTR)是一组利用个人病史为其定制治疗方案的决策规则。基于 Q-learning 的 Q-shared 算法已被用于制定涉及在多个干预阶段共享决策规则的 DTR。我们表明,由于在 Q-learning 设置中使用线性模型,现有 Q-shared 算法可能会出现不收敛,并确定了 Q-shared 失效的条件。我们开发了一种惩罚 Q-shared 算法,该算法不仅在违反该条件的设置中收敛,而且即使在条件满足时也能优于原始 Q-shared 算法。我们在一项真实世界应用和若干合成模拟中为所提出的方法提供了证据。
英文摘要:
A dynamic treatment regimen (DTR) is a set of decision rules to personalize treatments for an individual using their medical history. The Q-learning-based Q-shared algorithm has been used to develop DTRs that involve decision rules shared across multiple stages of intervention. We show that the existing Q-shared algorithm can suffer from non-convergence due to the use of linear models in the Q-learning setup, and identify the condition under which Q-shared fails. We develop a penalized Q-shared algorithm that not only converges in settings that violate the condition, but can outperform the original Q-shared algorithm even when the condition is satisfied. We give evidence for the proposed method in a real-world application and several synthetic simulations.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。