Bounding the Worst-class Error: A Boosting Approach
This addresses fairness and reliability issues in applications like medical diagnosis, where high error rates for critical classes are unacceptable, though it is an incremental improvement on existing methods for error minimization.
The paper tackles the problem of minimizing the worst-class error rate in classification, rather than the average error, by proposing a boosting approach that ensembles deep neural networks to bound this error and avoid overfitting, with experimental results showing reduced worst-class test error rates.
This paper tackles the problem of the worst-class error rate, instead of the standard error rate averaged over all classes. For example, a three-class classification task with class-wise error rates of 10%, 10%, and 40% has a worst-class error rate of 40%, whereas the average is 20% under the class-balanced condition. The worst-class error is important in many applications. For example, in a medical image classification task, it would not be acceptable for the malignant tumor class to have a 40% error rate, while the benign and healthy classes have a 10% error rates. To avoid overfitting in worst-class error minimization using Deep Neural Networks (DNNs), we design a problem formulation for bounding the worst-class error instead of achieving zero worst-class error. Moreover, to correctly bound the worst-class error, we propose a boosting approach which ensembles DNNs. We give training and generalization worst-class-error bound. Experimental results show that the algorithm lowers worst-class test error rates while avoiding overfitting to the training set. This code is available at https://github.com/saito-yuya/Bounding-the-Worst-class-error-A-Boosting-Approach.