Sharded Bayesian Additive Regression Trees
Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:BXVHNDVYrecord.jsonopen to challenge →
read the original abstract
In this paper we develop the randomized Sharded Bayesian Additive Regression Trees (SBT) model. We introduce a randomization auxiliary variable and a sharding tree to decide partitioning of data, and fit each partition component to a sub-model using Bayesian Additive Regression Tree (BART). By observing that the optimal design of a sharding tree can determine optimal sharding for sub-models on a product space, we introduce an intersection tree structure to completely specify both the sharding and modeling using only tree structures. In addition to experiments, we also derive the theoretical optimal weights for minimizing posterior contractions and prove the worst-case complexity of SBT.
This paper has not been read by Pith yet.
Forward citations
Cited by 2 Pith papers
-
Stabilizing the Splits through Minimax Decision Trees
MinimaxSplit trees and forests minimize maximum child impurity for splits, with oracle inequalities claiming faster excess risk convergence than CART under non-atomicity conditions, plus empirical gains on EEG regress...
-
Compactly-supported nonstationary kernels for computing exact Gaussian processes on big data
Derives compactly-supported nonstationary kernel enabling exact GPs on million-point datasets, shown on space-time temperature prediction.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.