# Very fast Bayesian Additive Regression Trees on GPU

arXiv: https://arxiv.org/abs/2410.23244 · HTML: https://mlsanity.com/paper/2410.23244/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 43 references that are in the map.
- First posted on arXiv 30 Oct 2024.

## Builds on (most cited first, 10 of 10)

- [XGBoost: A Scalable Tree Boosting System](https://mlsanity.com/paper/1603.02754/) (2016)
- [Automated versus do-it-yourself methods for causal inference: Lessons learned from a data analysis competition](https://mlsanity.com/paper/1707.02641/) (2017)
- [Bayesian Regression Tree Ensembles that Adapt to Smoothness and Sparsity](https://mlsanity.com/paper/1707.09461/) (2017)
- [bartMachine: Machine Learning with Bayesian Additive Regression Trees](https://mlsanity.com/paper/1312.2171/) (2013)
- [Atlantic Causal Inference Conference (ACIC) Data Analysis Challenge 2017](https://mlsanity.com/paper/1905.09515/) (2019)
- [Stochastic tree ensembles for regularized nonlinear regression](https://mlsanity.com/paper/2002.03375/) (2020)
- [A Mixing Time Lower Bound for a Simplified Version of BART](https://mlsanity.com/paper/2210.09352/) (2022)
- [XBART: Accelerated Bayesian Additive Regression Trees](https://mlsanity.com/paper/1810.02215/) (2018)
- [On the Gaussian process limit of Bayesian Additive Regression Trees](https://mlsanity.com/paper/2410.20289/) (2024)
- [On the Computational Efficiency of Bayesian Additive Regression Trees: An Asymptotic Analysis](https://mlsanity.com/paper/2406.19958/) (2024)

## Built on by (most cited first, 2 of 2)

- [BFTS: Thompson Sampling with Bayesian Additive Regression Trees](https://mlsanity.com/paper/2602.07767/) (2026)
- [A Ground-Truth Framework for Uncertainty Disentanglement with Posterior Risk](https://mlsanity.com/paper/2608.05995/) (2026)

## Abstract (arXiv)

Bayesian Additive Regression Trees (BART) is a nonparametric Bayesian regression technique based on an ensemble of decision trees. It is part of the toolbox of many statisticians. The overall statistical quality of the regression is typically higher than other generic alternatives, and it requires less manual tuning, making it a good default choice. However, it is a niche method compared to its natural competitor XGBoost, due to the longer running time, making sample sizes above 10,000-100,000 a nuisance. I present a GPU-enabled implementation of BART, faster by up to 200x relative to a single CPU core, making BART competitive in running time with XGBoost. This implementation is available in the Python package bartz.