Margin Distribution Controlled Boosting

TL;DR

This paper introduces MCBoost, a boosting algorithm that directly controls the margin distribution to improve generalization, demonstrating superior performance over existing methods on benchmark datasets.

Contribution

It proposes a novel boosting algorithm that explicitly optimizes margin distribution using a key adjustable parameter, with efficient implementation via column generation.

Findings

MCBoost outperforms AdaBoost, L2Boost, LPBoost, AdaBoost-CG, and MDBoost on UCI datasets.

AdaBoost is also a margin distribution controlled algorithm with iteration count influencing the distribution.

Direct control of margin distribution enhances boosting generalization performance.

Abstract

Schapire's margin theory provides a theoretical explanation to the success of boosting-type methods and manifests that a good margin distribution (MD) of training samples is essential for generalization. However the statement that a MD is good is vague, consequently, many recently developed algorithms try to generate a MD in their goodness senses for boosting generalization. Unlike their indirect control over MD, in this paper, we propose an alternative boosting algorithm termed Margin distribution Controlled Boosting (MCBoost) which directly controls the MD by introducing and optimizing a key adjustable margin parameter. MCBoost's optimization implementation adopts the column generation technique to ensure fast convergence and small number of weak classifiers involved in the final MCBooster. We empirically demonstrate: 1) AdaBoost is actually also a MD controlled algorithm and its iteration number acts as a parameter controlling the distribution and 2) the generalization performance of MCBoost evaluated on UCI benchmark datasets is validated better than those of AdaBoost, L2Boost, LPBoost, AdaBoost-CG and MDBoost.