On Computing Optimal Tree Ensembles

Authors: Christian Komusiewicz, Pascal Kunz, Frank Sommer, Manuel Sorge

ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical Mainly, we provide two novel algorithms for MTES and MMAXTES3 and matching lower-bound results for their running times. We call the first one witness-tree algorithm. This algorithm demonstrates that prospects for tractable algorithms for optimizing decision trees can be non-trivially generalized to optimizing tree ensembles.
Researcher Affiliation Academia 1Fakult at f ur Mathematik und Informatik, Friedrich-Schiller-Universit at Jena, Germany 2Algorithmics and Computational Complexity, TU Berlin, Germany 3Algorithm Engineering, HU Berlin, Germany 4Institute of Logic and Computation, TU Wien, Austria.
Pseudocode Yes Algorithm 1: Computing tree ensembles. 1 Function Refine Ensemble (C, (E, λ), S)
Open Source Code No No explicit statement or link indicating the availability of open-source code for the described methodology was found.
Open Datasets No The paper uses abstract training data sets (E, λ) and does not specify any named public datasets with access information or citations.
Dataset Splits No The paper discusses theoretical aspects and algorithms for classification but does not describe empirical experiments with training, validation, or test dataset splits.
Hardware Specification No The paper discusses theoretical algorithms and does not mention any specific hardware used for experiments.
Software Dependencies No The paper discusses theoretical algorithms and does not list any specific software dependencies with version numbers.
Experiment Setup No The paper focuses on theoretical algorithms and their complexity. It does not describe an experimental setup with hyperparameters or system-level training settings.