论文标题
G-NET：一种在动态治疗方面的反事实结果预测的深度学习方法
G-Net: A Deep Learning Approach to G-computation for Counterfactual Outcome Prediction Under Dynamic Treatment Regimes
论文作者
论文摘要
反事实预测是决策的基本任务。 G-Compunt是一种在动态时变治疗策略下估算预期的反事实结果的方法。现有的G-Compuntion实现主要采用具有捕获复杂时间和非线性依赖性结构能力有限的经典回归模型。本文介绍了G-NET，这是一个新型的G-Compuntion的新型深度学习框架，可以处理复杂的时间序列数据，同时施加最小的建模假设，并提供对个体或人群级别时间变化的治疗效果的估计。我们使用从CVSIM获得的现实复杂的时间模拟数据评估了替代G-NET实现，CVSIM是心血管系统的机械模型。
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.
