RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación
Artículo científico (JournalArticle)
EvoDeep: A new evolutionary approach for automatic Deep Neural Networks parametrisation
[EN] Deep Neural Networks (DNN) have become a powerful, and extremely popular mechanism, which has been widely used to solve problems of varied complexity, due to their ability to make models fitted to non-linear complex problems. Despite its well-known benefits, DNNs are complex learning models whose parametrisation and architecture are made usually by hand. This paper proposes a new Evolutionary Algorithm, named EvoDeep. devoted to evolve the parameters and the architecture of a DNN in order to maximise its classification accuracy, as well as maintaining a valid sequence of layers. This model is tested against a widely used dataset of handwritten digits images. The experiments performed using this dataset show that the Evolutionary Algorithm is able to select the parameters and the DNN architecture appropriately, achieving a 98.93% accuracy in the best run. (C) 2017 Elsevier Inc. All rights reserved., This work has been co-funded by the next research projects: EphemeCH (TIN2014-56494-C4-4-P) and DeepBio (TIN2017-85727-C4-3-P) Spanish Ministry of Economy and Competitivity and European Regional Development Fund FEDER, Justice Programme of the European Union (2014-2020) 723180 -RiskTrack-JUST-2015-JCOO-AG/JUST-2015-JCOO-AG-1, and by the CAM grant S2013/ICE-3095 (CIBERDINE:Cybersecurity, Data and Risks). The contents of this publication are the sole responsibility of their authors and can in no way be taken to reflect the views of the European Commission.