Online Multiclass Boosting
This work addresses the problem of extending boosting to online multiclass settings for machine learning practitioners, filling a theoretical and practical gap in the literature.
The authors tackled the gap in online multiclass boosting by defining a weak learning condition, leading to an optimal algorithm that minimizes the number of weak learners for a given accuracy, and proposed an adaptive near-optimal algorithm with strong real-data performance.
Recent work has extended the theoretical analysis of boosting algorithms to multiclass problems and to online settings. However, the multiclass extension is in the batch setting and the online extensions only consider binary classification. We fill this gap in the literature by defining, and justifying, a weak learning condition for online multiclass boosting. This condition leads to an optimal boosting algorithm that requires the minimal number of weak learners to achieve a certain accuracy. Additionally, we propose an adaptive algorithm which is near optimal and enjoys an excellent performance on real data due to its adaptive property.