Computational Eco-Systems for Handwritten Digits Recognition
Machine Learning
2017-05-29 v2
Abstract
Inspired by the importance of diversity in biological system, we built an heterogeneous system that could achieve this goal. Our architecture could be summarized in two basic steps. First, we generate a diverse set of classification hypothesis using both Convolutional Neural Networks, currently the state-of-the-art technique for this task, among with other traditional and innovative machine learning techniques. Then, we optimally combine them through Meta-Nets, a family of recently developed and performing ensemble methods.
Cite
@article{arxiv.1703.01872,
title = {Computational Eco-Systems for Handwritten Digits Recognition},
author = {Antonio Loquercio and Francesca Della Torre and Massimo Buscema},
journal= {arXiv preprint arXiv:1703.01872},
year = {2017}
}
Comments
Withdrawn from Journal for flaws in methodology