BAdaCost: Multi-class Boosting with Costs
This work addresses asymmetric multi-class classification problems, such as face and car detection, with an incremental improvement over existing methods.
The authors tackled multi-class cost-sensitive classification by introducing BAdaCost, a boosting algorithm that achieved significant performance gains compared to previous approaches, as demonstrated in experiments on face and car detection problems.
We present BAdaCost, a multi-class cost-sensitive classification algorithm. It combines a set of cost-sensitive multi-class weak learners to obtain a strong classification rule within the Boosting framework. To derive the algorithm we introduce CMEL, a Cost-sensitive Multi-class Exponential Loss that generalizes the losses optimized in various classification algorithms such as AdaBoost, SAMME, Cost-sensitive AdaBoost and PIBoost. Hence unifying them under a common theoretical framework. In the experiments performed we prove that BAdaCost achieves significant gains in performance when compared to previous multi-class cost-sensitive approaches. The advantages of the proposed algorithm in asymmetric multi-class classification are also evaluated in practical multi-view face and car detection problems.