A Bayesian Nonparametic Approach for Estimating Individualized Treatment-Response Curves

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Deep Counterfactual Networks with Propensity-Dropout (2017, methodology): “In particular, we compare the MSE (averaged over 100 experiments) achieved by the DCN-PD with those achieved by k nearest neighbor matching ( k -NN), Causal Forests with double-sample trees (Wager & Athey, 2015), Bayesian Additive Regression Trees (BART) (Chipman et al., 2010; Hill, 2012), and Bal-a”
Bayesian Inference of Individualized Treatment Effects using Multi-task Gaussian Processes (2017, methodology): “Most of the previous works estimate the ITE via co-variate adjustment (G-computation formula) [4, 5, 7, 11, 20]; the most remarkable of these methods are the nonparametric Bayesian additive regression trees [5] and causal forests [4, 9].”
Learning Treatment-Response Models from Multivariate Longitudinal Data (2017, methodology): “Recently, Xu et al. (2016) proposed a flexible Bayesian non-parametric approach for estimating univariate treatment response curves and predicting disease trajectories.”
What-If Reasoning with Counterfactual Gaussian Processes (2017, background): “Xu et al. [2016] also learn effects of actions in continuous-time on outcomes measured in continuous-time, but make one-step-ahead scalar predictions instead of trajectory-valued predictions.”