Optimal classification trees

• Published in: Machine learning Vol. 106; no. 7; pp. 1039 - 1082
• Main Authors: Bertsimas, Dimitris, Dunn, Jack
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.07.2017; Springer Nature B.V
• Subjects: Accuracy; Alignment; Artificial Intelligence; Axes (reference lines); Classification; Computer Science; Control; Datasets; Decision trees; Hardware; Heuristic; Heuristic methods; Hyperplanes; Machine learning; Mechatronics; Natural Language Processing (NLP); Optimization; Robotics; Simulation and Modeling; State of the art; Test procedures; Training; Decision trees; Mixed integer optimization; Classification
• ISSN: 0885-6125; 1573-0565
• DOI: 10.1007/s10994-017-5633-9

State-of-the-art decision tree methods apply heuristics recursively to create each split in isolation, which may not capture well the underlying characteristics of the dataset. The optimal decision tree problem attempts to resolve this by creating the entire decision tree at once to achieve global optimality. In the last 25 years, algorithmic advances in integer optimization coupled with hardware improvements have resulted in an astonishing 800 billion factor speedup in mixed-integer optimization (MIO). Motivated by this speedup, we present optimal classification trees , a novel formulation of the decision tree problem using modern MIO techniques that yields the optimal decision tree for axes-aligned splits. We also show the richness of this MIO formulation by adapting it to give optimal classification trees with hyperplanes that generates optimal decision trees with multivariate splits. Synthetic tests demonstrate that these methods recover the true decision tree more closely than heuristics, refuting the notion that optimal methods overfit the training data. We comprehensively benchmark these methods on a sample of 53 datasets from the UCI machine learning repository. We establish that these MIO methods are practically solvable on real-world datasets with sizes in the 1000s, and give average absolute improvements in out-of-sample accuracy over CART of 1–2 and 3–5% for the univariate and multivariate cases, respectively. Furthermore, we identify that optimal classification trees are likely to outperform CART by 1.2–1.3% in situations where the CART accuracy is high and we have sufficient training data, while the multivariate version outperforms CART by 4–7% when the CART accuracy or dimension of the dataset is low.