Multi-objective Evolution for Deep Neural Network Architecture Search

0534837 - ÚI 2021 RIV CH eng C - Conference Paper (international conference)
Vidnerová, Petra - Neruda, Roman
Multi-objective Evolution for Deep Neural Network Architecture Search.
Neural Information Processing. ICONIP 2020 Proceedings, Part III. Cham: Springer, 2020 - (Yang, H.; Pasupa, K.; Chi-Sing Leung, A.; Kwok, J.; Chan, J.; King, I.), s. 270-281. Lecture Notes on Computer Science, 12534. ISBN 978-3-030-63835-1. ISSN 0302-9743.
[ICONIP 2020: International Conference on Neural Information Processing /27./. Bangkok (TH), 23.11.2020-27.11.2020]
R&D Projects: GA ČR(CZ) GA18-23827S
Institutional support: RVO:67985807
Keywords : neural architecture search * deep neural networks * multi-objective evolution
OECD category: Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

In this paper, we propose a multi-objective evolutionary algorithm for automatic deep neural architecture search. The algorithm optimizes the performance of the model together with the number of network parameters. This allows exploring architectures that are both successful and compact. We test the proposed solution on several image classification data sets including MNIST, fashionMNIST and CIFAR-10, and we consider deep architectures including convolutional and fully connected networks. The effects of using two different versions of multi-objective selections are also examined in the paper. Our approach outperforms both the considered baseline architectures and the standard genetic algorithm used in our previous work.
Permanent Link: http://hdl.handle.net/11104/0312997