Recent work has extended the theoretical analysis of boosting algorithms to multiclass problems and online settings. However, the multiclass extension is in the batch setting and the online extensions only consider binary classification. To the best of our knowledge, there exists no framework to analyze online boosting algorithms for multiclass classification. We fill this gap in the literature by defining, and justifying, a weak learning condition for online multiclass boosting. We also provide an algorithm called online multiclass boost-by-majority to optimally combine weak learners in our setting.
View on arXiv