A Boosting Framework on Grounds of Online Learning

TL;DR

This paper introduces a boosting framework based on online learning duality, enabling the development of versatile algorithms for sparse boosting, smooth-distribution boosting, and agnostic learning, leveraging online learning insights.

Contribution

It presents a novel boosting framework rooted in online learning duality, facilitating the creation of multiple new algorithms for diverse learning scenarios.

Findings

Developed algorithms for sparse boosting and smooth-distribution boosting.

Extended the framework to agnostic learning and double-projection online algorithms.

Demonstrated the framework's effectiveness through theoretical analysis.

Abstract

By exploiting the duality between boosting and online learning, we present a boosting framework which proves to be extremely powerful thanks to employing the vast knowledge available in the online learning area. Using this framework, we develop various algorithms to address multiple practically and theoretically interesting questions including sparse boosting, smooth-distribution boosting, agnostic learning and some generalization to double-projection online learning algorithms, as a by-product.