Research

Paper

TESTING February 25, 2026

LoBoost: Fast Model-Native Local Conformal Prediction for Gradient-Boosted Trees

Authors

Vagner Santos, Victor Coscrato, Luben Cabezas, Rafael Izbicki, Thiago Ramos

Abstract

Gradient-boosted decision trees are among the strongest off-the-shelf predictors for tabular regression, but point predictions alone do not quantify uncertainty. Conformal prediction provides distribution-free marginal coverage, yet split conformal uses a single global residual quantile and can be poorly adaptive under heteroscedasticity. Methods that improve adaptivity typically fit auxiliary nuisance models or introduce additional data splits/partitions to learn the conformal score, increasing cost and reducing data efficiency. We propose LoBoost, a model-native local conformal method that reuses the fitted ensemble's leaf structure to define multiscale calibration groups. Each input is encoded by its sequence of visited leaves; at resolution level k, we group points by matching prefixes of leaf indices across the first k trees and calibrate residual quantiles within each group. LoBoost requires no retraining, auxiliary models, or extra splitting beyond the standard train/calibration split. Experiments show competitive interval quality, improved test MSE on most datasets, and large calibration speedups.

Metadata

arXiv ID: 2602.22432
Provider: ARXIV
Primary Category: stat.ML
Published: 2026-02-25
Fetched: 2026-02-27 04:35

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.22432v1</id>\n    <title>LoBoost: Fast Model-Native Local Conformal Prediction for Gradient-Boosted Trees</title>\n    <updated>2026-02-25T21:44:19Z</updated>\n    <link href='https://arxiv.org/abs/2602.22432v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.22432v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Gradient-boosted decision trees are among the strongest off-the-shelf predictors for tabular regression, but point predictions alone do not quantify uncertainty. Conformal prediction provides distribution-free marginal coverage, yet split conformal uses a single global residual quantile and can be poorly adaptive under heteroscedasticity. Methods that improve adaptivity typically fit auxiliary nuisance models or introduce additional data splits/partitions to learn the conformal score, increasing cost and reducing data efficiency. We propose LoBoost, a model-native local conformal method that reuses the fitted ensemble's leaf structure to define multiscale calibration groups. Each input is encoded by its sequence of visited leaves; at resolution level k, we group points by matching prefixes of leaf indices across the first k trees and calibrate residual quantiles within each group. LoBoost requires no retraining, auxiliary models, or extra splitting beyond the standard train/calibration split. Experiments show competitive interval quality, improved test MSE on most datasets, and large calibration speedups.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='stat.ML'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n    <published>2026-02-25T21:44:19Z</published>\n    <arxiv:primary_category term='stat.ML'/>\n    <author>\n      <name>Vagner Santos</name>\n    </author>\n    <author>\n      <name>Victor Coscrato</name>\n    </author>\n    <author>\n      <name>Luben Cabezas</name>\n    </author>\n    <author>\n      <name>Rafael Izbicki</name>\n    </author>\n    <author>\n      <name>Thiago Ramos</name>\n    </author>\n  </entry>"
}