Skip to content
DraconiticResearch
Search Sources API
201,683 studies31 sourcesUpdated 47 min ago
PreprintarXiv licencearXiv

Improving the Predictive Performance of Bootstrap Aggregating by Dirichlet Resampling

  • Quoc Viet Le
  • Joonha Park
Open at sourcePDFDOICite

Abstract

We revisit Breiman's observation that reducing inter-tree correlation without weakening individual trees can improve random forests. Building on this principle, we introduce two variants: Dirichlet-Multinomial Bagging Random Forest (DM) and Dirichlet-Weighted Random Forest (DW). Both modulate sample reweighting via a concentration parameter α>0. We provide a simple theoretical criterion that clarifies when these variants behave indistinguishably from standard random forests, and we use it to guide a lightweight tuning strategy. In a controlled evaluation on public classification benchmarks, DM and DW are consistently competitive and often stronger than other random-forest (RF) baselines, with negligible additional runtime.

Related

  1. A Dirichlet-Quantile Bootstrap approach for the probabilistic resampling of time series
    Ahmed Hamimes, Abdennour Boulesnane
    arXivPreprintCC BY-NC-ND 4.0Full textStatisticsarXiv
  2. Estimating Stability of Individual-Specific Predictions from AI-based prediction models using Bootstrap and Cross-validation Resampling
    Alex Carriero, Martijn L. G. van Dam, Anne A. H. de Hond +5
    arXivPreprintCC BY-NC-ND 4.0Full textStatisticsarXiv
  3. Bagged Martingale Posteriors: Calibrated Uncertainty Quantification for Predictive Resampling
    Hui Wang, Edwin Fong, David T. Frazier
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  4. Improving the performance of regularized Stokeslet simulations using Rosenbrock time integration methods
    Moslem Uddin, Lisa J. Fauci, Ricardo Cortez +1
    arXivPreprintCC BY 4.0Full textNumerical analysis & scientific computingBiological & medical physicsarXiv
  5. Improving sport horse selection using machine learning: Early prediction of elite-level show jumping performance
    Marco Zanchi, Clara Bordin, Michela Ablondi +4
    PLOS OnePeer-reviewedCC BY 4.0Full textZoology & animal behaviourCombinatorics & discrete mathematicsPLOS

Cite

BibTeXRIS
BibTeX
@misc{le2026improving,
  title = {{Improving the Predictive Performance of Bootstrap Aggregating by Dirichlet Resampling}},
  author = {Quoc Viet Le and Joonha Park},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.21454},
  eprint = {2609.21454},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.21454},
}
RIS
TY  - UNPB
TI  - Improving the Predictive Performance of Bootstrap Aggregating by Dirichlet Resampling
AU  - Quoc Viet Le
AU  - Joonha Park
PY  - 2026
DA  - 2026/09/18
PB  - arXiv
DO  - 10.48550/arxiv.2609.21454
UR  - https://arxiv.org/abs/2609.21454
AB  - We revisit Breiman's observation that reducing inter-tree correlation without weakening individual trees can improve random forests. Building on this principle, we introduce two variants: Dirichlet-Multinomial Bagging Random Forest (DM) and Dirichlet-Weighted Random Forest (DW). Both modulate sample reweighting via a concentration parameter α>0. We provide a simple theoretical criterion that clarifies when these variants behave indistinguishably from standard random forests, and we use it to guide a lightweight tuning strategy. In a controlled evaluation on public classification benchmarks, DM and DW are consistently competitive and often stronger than other random-forest (RF) baselines, with negligible additional runtime.
ER  -