MNIST-NET10: A heterogeneous deep networks fusion based on the degree of certainty to reach 0.1 error rate. Ensembles overview and proposal
This work addresses the challenge of enhancing ensemble methods for image classification, but it is incremental as it builds on existing fusion schemes.
The paper tackled the problem of improving classification accuracy on MNIST by proposing MNIST-NET10, a complex heterogeneous fusion architecture based on a degree of certainty aggregation approach, which achieved a new record with only 10 misclassified images.
Ensemble methods have been widely used for improving the results of the best single classificationmodel. A large body of works have achieved better performance mainly by applying one specific ensemble method. However, very few works have explored complex fusion schemes using het-erogeneous ensembles with new aggregation strategies. This paper is three-fold: 1) It provides an overview of the most popular ensemble methods, 2) analyzes several fusion schemes using MNIST as guiding thread and 3) introduces MNIST-NET10, a complex heterogeneous fusion architecture based on a degree of certainty aggregation approach; it combines two heterogeneous schemes from the perspective of data, model and fusion strategy. MNIST-NET10 reaches a new record in MNISTwith only 10 misclassified images. Our analysis shows that such complex heterogeneous fusionarchitectures based on the degree of certainty can be considered as a way of taking benefit fromdiversity.