Classy Ensemble: A Novel Ensemble Algorithm for Classification

We present Classy Ensemble, a novel ensemble-generation algorithm for classification tasks, which aggregates models through a weighted combination of per-class accuracy. Tested over 153 machine learning datasets we demonstrate that Classy Ensemble outperforms two other well-known aggregation algorithms -- order-based pruning and clustering-based pruning -- as well as the recently introduced lexigarden ensemble generator. Classy Ensemble also fares favorably with deep networks, over four image datasets: Fashion MNIST, CIFAR10, CIFAR100, and ImageNet.