Best-scored Random Forest Classification

TL;DR

This paper introduces a best-scored random forest algorithm that selects the best-performing trees from random candidates, achieving higher accuracy and theoretical optimality in binary classification tasks.

Contribution

It proposes a novel best-scored selection method for random forests, with theoretical convergence guarantees and practical efficiency improvements.

Findings

Achieves higher accuracy than traditional random forests

Establishes almost optimal convergence rates under certain conditions

Demonstrates superior performance in numerical experiments

Abstract

We propose an algorithm named best-scored random forest for binary classification problems. The terminology "best-scored" means to select the one with the best empirical performance out of a certain number of purely random tree candidates as each single tree in the forest. In this way, the resulting forest can be more accurate than the original purely random forest. From the theoretical perspective, within the framework of regularized empirical risk minimization penalized on the number of splits, we establish almost optimal convergence rates for the proposed best-scored random trees under certain conditions which can be extended to the best-scored random forest. In addition, we present a counterexample to illustrate that in order to ensure the consistency of the forest, every dimension must have the chance to be split. In the numerical experiments, for the sake of efficiency, we employ an adaptive random splitting criterion. Comparative experiments with other state-of-art classification methods demonstrate the accuracy of our best-scored random forest.