AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

VaRT: Variational Regression Trees

Sebastian Salazar

NeurIPS 2023 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Nonparametric and PAC-Bayes Bayesian learning · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace BART (50 trees) with VaRT for ATE estimationVaRT · IHDP linear response surfaceATE estimate: 3.6 -> 3.86Improved
replace BART (50 trees) with VaRT for ATE estimationVaRT · IHDP non-linear response surfaceATE estimate: 2.9 -> 3.80Improved
increase VaRT tree depth from 3 to 10 (ablation)VaRT · concreteslumpRMSE: 0.009 -> 0.047Hurt
increase VaRT tree depth from 3 to 10 (ablation)VaRT · autompgRMSE: 0.116 -> 0.109Improved
increase VaRT tree depth from 3 to 10 (ablation)VaRT · airfoilRMSE: 0.151 -> 0.131Improved
increase VaRT tree depth from 3 to 10 (ablation)VaRT · housingRMSE: 0.101 -> 0.101No clear effect
single VaRT tree instead of 50-tree CatBoost ensembleVaRT3 · concreteslumpRMSE: 0.209 -> 0.009Improved
single VaRT tree instead of 50-tree XGBoost ensembleVaRT7 · airfoilRMSE: 0.061 -> 0.131Hurt
single VaRT tree instead of 50-tree RandomForestVaRT5 · housingRMSE: 0.118 -> 0.097Improved
VaRT depth 5 versus CatBoost on a large datasetVaRT · tamielectric (45781 datapoints)RMSE: 0.503 -> 0.508No clear effect
VaRT depth 5 versus XGBoost on a large datasetVaRT · kin40k (40000 datapoints)RMSE: 0.110 -> 0.099Improved
VaRT (median) versus BART with 50 treesVaRT · autosRMSE: 0.075 -> 0.107Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy