Unsupervised Estimation of Ensemble Accuracy
This work addresses the challenge of evaluating ensemble performance in unsupervised learning scenarios, which is incremental as it builds on existing diversity measures by removing the need for labels.
The paper tackles the problem of estimating ensemble classifier accuracy without labeled data, presenting a method based on a combinatorial bound that efficiently predicts misclassifications. It demonstrates the method on large-scale face recognition datasets, showing practical utility in unsupervised settings.
Ensemble learning combines several individual models to obtain a better generalization performance. In this work we present a practical method for estimating the joint power of several classifiers. It differs from existing approaches which focus on "diversity" measures by not relying on labels. This makes it both accurate and practical in the modern setting of unsupervised learning with huge datasets. The heart of the method is a combinatorial bound on the number of mistakes the ensemble is likely to make. The bound can be efficiently approximated in time linear in the number of samples. We relate the bound to actual misclassifications, hence its usefulness as a predictor of performance. We demonstrate the method on popular large-scale face recognition datasets which provide a useful playground for fine-grain classification tasks using noisy data over many classes.