Fast-DENSER: Fast Deep Evolutionary Network Structured Representation

• Main Authors: Filipe Assunção, Nuno Lourenço, Bernardete Ribeiro, Penousal Machado
• Format: Article
• Language: English
• Published: Elsevier 2021-06-01
• Series: SoftwareX
• Subjects: Artificial Neural Networks; Automated machine learning; NeuroEvolution
• Online Access: http://www.sciencedirect.com/science/article/pii/S235271102100039X

This paper introduces a grammar-based general purpose framework for the automatic search and deployment of potentially Deep Artificial Neural Networks (DANNs). The approach is known as Fast Deep Evolutionary Network Structured Representation (Fast-DENSER) and is capable of simultaneously optimising the topology, learning strategy and any other required hyper-parameters (e.g., data pre-processing or augmentation). Fast-DENSER has been successfully applied to numerous object recognition tasks, with the generation of Convolutional Neural Networks (CNNs). The code is developed and tested in Python3, and made available as a library. A simple and easy to follow example is described for the automatic search of CNNs for the Fashion-MNIST benchmark.