Regularized impurity reduction: accurate decision trees with complexity guarantees
Open Access
- 28 November 2022
- journal article
- research article
- Published by Springer Nature in Data Mining and Knowledge Discovery
- Vol. 37 (1) , 434-475
Abstract
Decision trees are popular classification models, providing high accuracy and intuitive explanations. However, as the tree size grows the model interpretability deteriorates. Traditional tree-induction algorithms, such as C4.5 and CART, rely on impurity-reduction functions that promote the discriminative power of each split. Thus, although these traditional methods are accurate in practice, there has been no theoretical guarantee that they will produce small trees. In this paper, we justify the use of a general family of impurity functions, including the popular functions of entropy and Gini-index, in scenarios where small trees are desirable, by showing that a simple enhancement can equip them with complexity guarantees. We consider a general setting, where objects to be classified are drawn from an arbitrary probability distribution, classification can be binary or multi-class, and splitting tests are associated with non-uniform costs. As a measure of tree complexity, we adopt the expected cost to classify an object drawn from the input distribution, which, in the uniform-cost case, is the expected number of tests. We propose a tree-induction algorithm that gives a logarithmic approximation guarantee on the tree complexity. This approximation factor is tight up to a constant factor under mild assumptions. The algorithm recursively selects a test that maximizes a greedy criterion defined as a weighted sum of three components. The first two components encourage the selection of tests that improve the balance and the cost-efficiency of the tree, respectively, while the third impurity-reduction component encourages the selection of more discriminative tests. As shown in our empirical evaluation, compared to the original heuristics, the enhanced algorithms strike an excellent balance between predictive accuracy and tree complexity.Keywords
All Related Versions
Funding Information
- Research Council of Finland (317085, 325117)
- European Research Council (834862)
- Horizon 2020 (871042)
- Wallenberg Ai, Autonomous Systems and Software Program
This publication has 21 references indexed in Scilit:
- Minimum Latency Submodular CoverLecture Notes in Computer Science, 2012
- Average-Case Active Learning with Costs2009
- Approximating Optimal Binary Decision Trees2008
- Approximating Min Sum Set CoverAlgorithmica, 2004
- On the Boosting Ability of Top–Down Decision Tree Learning AlgorithmsJournal of Computer and System Sciences, 1999
- On an Optimal Split Tree Problem1999
- A threshold of ln n for approximating set coverJournal of the ACM, 1998
- Lower Bounds on Learning Decision Lists and TreesInformation and Computation, 1996
- Constructing optimal binary decision trees is NP-completeInformation Processing Letters, 1976
- Optimal Binary Identification ProceduresSIAM Journal on Applied Mathematics, 1972