arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2003.10551cs.LGstat.ML
收藏

G-Net:一种用于动态治疗方案下的伪事实结果预测的G计算的深度学习方法

G-Net: A Deep Learning Approach to G-computation for Counterfactual Outcome Prediction Under Dynamic Treatment Regimes

Massachusetts Institute of Technology(麻省理工学院);IBM Research(IBM研究院);MIT-IBM Watson AI Lab(麻省理工学院-IBM沃森人工智能实验室)

Rui Li, Zach Shahn, Jun Li, Mingyu Lu, Prithwish Chakraborty, Daby Sow, Mohamed Ghalwash, Li-wei H. Lehman

摘要:

反事实预测是决策中的一项基本任务。G计算是一种在动态时变治疗策略下估计预期反事实结果的方法。现有的G-计算实现大多采用经典的回归模型,其捕获复杂的时间和非线性依赖结构的能力有限。本文介绍了G-Net,这是一种用于G计算的新型顺序深度学习框架,可以处理复杂的时间序列数据,同时施加最小的建模假设,并提供对个体或群体水平时变治疗效果的估计。我们使用从CVSim(一种心血管系统的机械模型)获得的逼真的复杂时间模拟数据来评估替代G-Net实现。

英文摘要:

Counterfactual prediction is a fundamental task in decision-making. G-computation is a method for estimating expected counterfactual outcomes under dynamic time-varying treatment strategies. Existing G-computation implementations have mostly employed classical regression models with limited capacity to capture complex temporal and nonlinear dependence structures. This paper introduces G-Net, a novel sequential deep learning framework for G-computation that can handle complex time series data while imposing minimal modeling assumptions and provide estimates of individual or population-level time varying treatment effects. We evaluate alternative G-Net implementations using realistically complex temporal simulated data obtained from CVSim, a mechanistic model of the cardiovascular system.

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。

↑