Bayesian Additive Regression Trees

2006

331 citations recorded by the source corpus.

Subfields: Probabilistic & Bayesian Methods

Citations in the corpus

• All Models are Wrong, but Many are Useful: Learning a Variable’s Importance by Studying an Entire Class of Prediction Models Simultaneously (2018, neutral): “…model classes where profile likelihoods are difficult to compute, such as neural networks or random forests, approximate inference is sometimes achieved via approximate optimization procedures (for example, Markov chain Monte Carlo for Bayesian additive regression trees, in Chipman et al., 2010).”
• Deep Counterfactual Networks with Propensity-Dropout (2017, neutral): “…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-ancing neural networks (BNN) (Johansson et al., 2016).”
• Learning Representations for Counterfactual Inference (2016, positive): “The performance of BART on News is likely hurt by the dimensionality of the dataset, and could improve with hyperparameter tuning.”
• From Real‐World Patient Data to Individualized Treatment Effects Using Machine Learning: Current and Future Methods to Address Underlying Challenges (2020, unknown): “61 evaluate Bayesian Additive Regression Trees 52,53 and Causal Forest 54,55 on estimating the impact of the volume of transcatheter aortic valve replacement on 30-day hospital readmission, 62 as well as the effect of endovascular and surgical revascularization therapy for critical limb ischemia on…”
• Bayesian and Empirical Bayesian Forests (2015, unknown): “As an alternative, the original BCART authors developed the Bayesian Additive Regression Trees (BART; Chipman et al., 2010) algorithm, which replaces a single tree parameter target with the sum of many small trees.”
• : "Statistical Modeling: The Two Cultures" (Statistical Science, 2001, 16(3), 199-231) (unknown): “It has made strides in understanding theoretical aspects of random forests, (Scornet et al., 2015; Wager and Walther, 2015; Athey et al., 2019) and boosting (B¨uhlmann and Yu, 2003; Zhang and Yu, 2005), in understanding prediction accuracy from scratch by establishing non-asymptotic prediction error”
• Ensemble Trees and CLTs: Statistical Inference for Supervised Learning (2014, unknown): “Chipman et al. [2010] have received significant attention for developing BART , a Bayesian “sum-of-trees” statistical model for the underlying regression function that allows for pointwise posterior inference throughout the feature space as well as estimates for individual feature effects.”
• Adaptive Hyper-box Matching for Interpretable Individualized Treatment Effect Estimation (2020, unknown): “We compare the following estimators: BART - Bayesian Additive Regression Trees (Chipman et al. 2010; J. L. Hill 2011) estimates ITEi as f̂1(xi) − f̂0(xi), Best CF - 1 : k matching of a treated unit i to the k control units with outcomes closest to i’s true counterfactual (this is cheating: one does…”