AnyDTree: An Anytime Solver for Perfect Decision Trees

Finding progressively smaller trees with 100% training accuracy

Bachelor Thesis (2025)
Author(s)

I. Hosu (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Contributor(s)

Jacobus G.M. van der Linden – Mentor (TU Delft - Algorithmics)

Emir Demirovic – Mentor (TU Delft - Algorithmics)

Daniël Vos – Mentor (TU Delft - Algorithmics)

Jasmijn A. Baaijens – Graduation committee member (TU Delft - Pattern Recognition and Bioinformatics)

Faculty
Electrical Engineering, Mathematics and Computer Science
More Info
expand_more
Publication Year
2025
Language
English
Graduation Date
27-06-2025
Awarding Institution
Delft University of Technology
Project
['CSE3000 Research Project']
Programme
['Computer Science and Engineering']
Faculty
Electrical Engineering, Mathematics and Computer Science
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Finding the smallest decision tree that perfectly fits the training data is NP-complete; yet, such trees remain attractive due to their interpretability and minimal footprint. Existing solutions occupy two extremes: heuristics like CART instantly produce trees but remain far from optimal, whereas exact solvers like Witty give no intermediate output. We introduce AnyDTree, an anytime algorithm that continuously maintains a 100% accurate tree and monotonically shrinks its size. It employs an expand-and-backtrack search that ensures complete solutions at every step, combined with aggressive pruning and caching mechanisms to eliminate redundant exploration. On 70 binary-classification variants of 35 UCI datasets, AnyDTree incurs no statistically significant overhead in finding the optimal size compared with Witty (log-rank p>0.1). On the 46 datasets with known optima, it demonstrates significantly improved anytime behaviour - measured by the confined primal integral - with a median score of 0.00034, outperforming both Witty (0.00059) and CART (0.20) (p<0.001). These results position AnyDTree as a practical middle ground between heuristic and exact solutions.

Files

AnyDTree_-_Iulia_Hosu.pdf
(pdf | 1.66 Mb)
License info not available