arXiv Statistics ML
Seeing the Forest for the Trees: The Gaussian Process Limit of BART
Read the original on arXiv Statistics ML →The Flow has not summarised this story yet — read it at arXiv Statistics ML.
Related stories
arXiv Machine Learning
Oct 8
RoBART: Bayesian Additive Regression Trees with Tree-Specific Rotations
arXiv:2610.10214v1 Announce Type: cross Abstract: Bayesian additive regression trees (BART) can require many splits to approximate boundaries misaligned with the predictor axes. RoBART assigns each t...
arXiv Machine Learning
Sep 24
Dirichlet Process Mixtures of Trees with Gaussian Process Splits: A Bayesian Nonparametric Framework with Posterior Contraction Rate
arXiv:2609. 27930v1 Announce Type: cross Abstract: We propose a Bayesian nonparametric mixture of regression trees with a Dirichlet process prior over tree-parameter pairs, enabling data-driven selection of ensemble size and unifying CART, BART, random forests, and boosting.
Hugging Face Trending Papers
Oct 7
RoBART: Bayesian Additive Regression Trees with Tree-Specific Rotations
Bayesian additive regression trees (BART) can require many splits to approximate boundaries misaligned with the predictor axes. RoBART assigns each tree a rotation shared by all internal nodes, retain...
arXiv Machine Learning
Aug 27
Fast rates in Bayesian online learning with approximate posteriors
The paper investigates how fast predictive regret guarantees of exact Bayesian online learning can be maintained when using approximate posterior methods. It establishes a general theorem linking the cumulative cost of posterior approximation to the contraction radius of the exact Gibbs posterior and the Wasserstein distance between approximate and exact posteriors. Three concrete online learning scenarios—linear models, infinite‑dimensional exponential families, and Gaussian process regression—illustrate that appropriately accurate approximations (projected Langevin, truncation, and sparse variational posteriors) preserve fast regret bounds while reducing computational demands.
arXiv Machine Learning
Jun 30
Warm-Starting Iterative Gaussian Processes for Faster Sequential Inference
arXiv:2511. 16340v2 Announce Type: replace Abstract: Efficient Gaussian process (GP) inference is critical for sequential decision-making tasks such as active learning, online prediction, and Bayesian optimization.
arXiv Statistics ML
Oct 5
A Residual Tree Gaussian Process Modeling Framework for High-Dimensional Data
The paper introduces ResTGP, a Bayesian residual tree Gaussian Process framework designed for large, high‑dimensional spatial datasets with heterogeneous structures. It decomposes a Gaussian process across a dyadic tree, iteratively computing predictive and residual processes at each node to capture multi‑scale covariance efficiently. The method employs recursive message passing for linear‑time Bayesian inference and demonstrates posterior consistency, with numerical experiments and a storm surge application highlighting its advantages.