A Boosting Framework on Grounds of Online Learning
This work addresses boosting challenges in machine learning, offering a novel framework that integrates online learning insights, though it appears incremental in its application to existing problems.
The authors tackled the problem of boosting by leveraging its duality with online learning, resulting in a powerful framework that enabled the development of algorithms for sparse boosting, smooth-distribution boosting, agnostic learning, and generalizations to double-projection online learning.
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.