Interpretable Random Forests via Rule Extraction

TL;DR

SIRUS is a novel rule-based regression algorithm that offers a stable, interpretable, and accurate alternative to traditional black-box models like random forests, with proven asymptotic stability and practical software implementation.

Contribution

This paper introduces SIRUS, a stable and interpretable rule extraction method from random forests, combining simplicity, stability, and predictive accuracy.

Findings

SIRUS maintains high predictive accuracy comparable to random forests.

SIRUS demonstrates asymptotic stability both empirically and theoretically.

The software implementation is available on CRAN.

Abstract

We introduce SIRUS (Stable and Interpretable RUle Set) for regression, a stable rule learning algorithm which takes the form of a short and simple list of rules. State-of-the-art learning algorithms are often referred to as "black boxes" because of the high number of operations involved in their prediction process. Despite their powerful predictivity, this lack of interpretability may be highly restrictive for applications with critical decisions at stake. On the other hand, algorithms with a simple structure-typically decision trees, rule algorithms, or sparse linear models-are well known for their instability. This undesirable feature makes the conclusions of the data analysis unreliable and turns out to be a strong operational limitation. This motivates the design of SIRUS, which combines a simple structure with a remarkable stable behavior when data is perturbed. The algorithm is based on random forests, the predictive accuracy of which is preserved. We demonstrate the efficiency of the method both empirically (through experiments) and theoretically (with the proof of its asymptotic stability). Our R/C++ software implementation sirus is available from CRAN.