arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2004.02778stat.MLcs.LG
收藏

评论:动态治疗方案的熵学习

Comment: Entropy Learning for Dynamic Treatment Regimes

Cornell University(康奈尔大学)

Nathan Kallus

摘要:

祝贺教授。Binyan Jiang,Rui Song,Jialiang Li,and Donglin Zeng(JSLZ)在使用熵损失作为替代对通过经验风险最小化学习的最佳动态治疗方案(DTR)进行推断方面取得了令人兴奋的进展。JSLZ的方法利用基于逆概率加权(IPW)的给定决策规则的值的拒绝和重要性抽样估计及其作为加权(或成本敏感)分类的解释。他们使用平滑的分类替代,使他们能够仔细的方法来分析渐近分布。然而,即使是出于评估目的,IPW估计也是有问题的,因为它导致权重丢弃了大部分数据,并且在剩余的数据上变化非常大。在这篇评论中,我将讨论一个基于优化的替代方法来评估DTR,回顾几个连接,并提出前进的方向。这将Kallus(2018a)的平衡政策评估方法扩展到纵向环境。

英文摘要:

I congratulate Profs. Binyan Jiang, Rui Song, Jialiang Li, and Donglin Zeng (JSLZ) for an exciting development in conducting inferences on optimal dynamic treatment regimes (DTRs) learned via empirical risk minimization using the entropy loss as a surrogate. JSLZ's approach leverages a rejection-and-importance-sampling estimate of the value of a given decision rule based on inverse probability weighting (IPW) and its interpretation as a weighted (or cost-sensitive) classification. Their use of smooth classification surrogates enables their careful approach to analyzing asymptotic distributions. However, even for evaluation purposes, the IPW estimate is problematic as it leads to weights that discard most of the data and are extremely variable on whatever remains. In this comment, I discuss an optimization-based alternative to evaluating DTRs, review several connections, and suggest directions forward. This extends the balanced policy evaluation approach of Kallus (2018a) to the longitudinal setting.

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。

↑