Skip to main content

MONCAE: Multi-Objective Neuroevolution of Convolutional Autoencoders.

Dimanov, D., Balaguer-Ballester, E., Rostami, S. and Singleton, C., 2021. MONCAE: Multi-Objective Neuroevolution of Convolutional Autoencoders. In: ICLR2021: 2nd Workshop on Neural Architecture Search (NAS 21), 7 May 2021, Online.

Full text available as:

[img]
Preview
PDF
MONCAE-2.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial.

801kB

Abstract

In this paper, we present a novel neuroevolutionary method to identify the architecture and hyperparameters of convolutional autoencoders. Remarkably, we used a hypervolume indicator in the context of neural architecture search for autoencoders, for the first time to our current knowledge. Results show that images were compressed by a factor of more than 10, while still retaining enough information to achieve image classification for the majority of the tasks. Thus, this new approach can be used to speed up the AutoML pipeline for image compression.

Item Type:Conference or Workshop Item (Paper)
Group:Faculty of Science & Technology
ID Code:35800
Deposited By: Unnamed user with email symplectic@symplectic
Deposited On:20 Jul 2021 13:01
Last Modified:20 Jul 2021 13:01

Downloads

Downloads per month over past year

More statistics for this item...
Repository Staff Only -