Automating autoencoder architecture configuration: An evolutionary approach

• Francisco Charte ^[1] ; Antonio Rivera ^[1] ; Francisco Martínez ^[1] ; María del Jesus ^[1]
  1. [1] Universidad de Jaén

     Universidad de Jaén

     Jaén, España

     
• Localización: Understanding the Brain Function and Emotions: 8th International Work-Conference on the Interplay Between Natural and Artificial Computation, IWINAC 2019 Almería, Spain, June 3–7, 2019 Proceedings, Part I / José Manuel Ferrández Vicente (dir. congr.), José Ramón Álvarez Sánchez (dir. congr.), Félix de la Paz López (dir. congr.), Francisco Javier Toledo Moreo (dir. congr.), Hojjat Adeli (dir. congr.), 2019, ISBN 978-3-030-19591-5, págs. 339-349
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • Learning from existing data allows building models able toclassify patterns, infer association rules, predict future values in time series and much more. Choosing the right features is a vital step of the learning process, specially while dealing with high-dimensional spaces.Autoencoders (AEs) have shown ability to conduct manifold learning, compressing the original feature space without losing useful information.However, there is no optimal AE architecture for all datasets. In thispaper we show how to use evolutionary approaches to automate AE architecture configuration. First, a coding to embed the AE configuration in a chromosome is proposed.Then, two evolutionary alternatives are compared against exhaustive search. The results show the great superiority of the evolutionary way.