A more robust boosting algorithm

TL;DR

This paper introduces a new boosting algorithm inspired by large margins theory, demonstrating increased robustness to label noise through experimental validation.

Contribution

The paper proposes a novel boosting algorithm that improves robustness to label noise, advancing the state of boosting methods.

Findings

Significantly more robust against label noise

Experimental evidence supports robustness claims

Based on large margins theory

Abstract

We present a new boosting algorithm, motivated by the large margins theory for boosting. We give experimental evidence that the new algorithm is significantly more robust against label noise than existing boosting algorithm.