Factorized MultiClass Boosting
This addresses multiclass classification efficiency for users needing faster training times, though it appears incremental as it builds on existing regression and tree methods.
The paper tackles multiclass classification by decomposing it into regression tasks solved with CART trees, achieving the same model quality as state-of-the-art solutions but significantly faster, with robustness to imbalanced datasets without re-balancing.
In this paper, we introduce a new approach to multiclass classification problem. We decompose the problem into a series of regression tasks, that are solved with CART trees. The proposed method works significantly faster than state-of-the-art solutions while giving the same level of model quality. The algorithm is also robust to imbalanced datasets, allowing to reach high-quality results in significantly less time without class re-balancing.