论文标题
学习重叠表示以估计个性化治疗效果
Learning Overlapping Representations for the Estimation of Individualized Treatment Effects
论文作者
论文摘要
与替代方案相比，进行干预的选择取决于其潜在益处或危害。从未观察到所有结果，从观察数据中估算替代结果的可能结果是一个具有挑战性的问题，并且选择偏见排除了不同介绍组的直接比较。尽管取得了经验的成功，但我们表明，学习域的不变性表示输入的算法（对预测进行预测）通常是不合适的，并发展了概括范围，以证明对域重叠的依赖性并突出了对不可逆转潜在潜伏地图的需求。基于这些结果，我们开发了一种深内核回归算法和后正规化框架，该框架在各种基准数据集上大大优于最先进的框架。
The choice of making an intervention depends on its potential benefit or harm in comparison to alternatives. Estimating the likely outcome of alternatives from observational data is a challenging problem as all outcomes are never observed, and selection bias precludes the direct comparison of differently intervened groups. Despite their empirical success, we show that algorithms that learn domain-invariant representations of inputs (on which to make predictions) are often inappropriate, and develop generalization bounds that demonstrate the dependence on domain overlap and highlight the need for invertible latent maps. Based on these results, we develop a deep kernel regression algorithm and posterior regularization framework that substantially outperforms the state-of-the-art on a variety of benchmarks data sets.
