Title: Optimisation of spatial-exploitation CNN models through hyperparameter-tuning and human-in-the-loop combination

Authors: Luke Beveridge; Keshav Dahal

Addresses: School of Computing Engineering and Physical Sciences, University of the West of Scotland, Paisley, PA1 2BE, UK ' School of Computing Engineering and Physical Sciences, University of the West of Scotland, Paisley, PA1 2BE, UK

Abstract: Spatial-exploitation convolutional neural networks (CNNs) have a simplified architecture compared to other CNN models. However, devices with limited computational resources could struggle with processing spatial-exploitation CNNs. To address this, we investigate two methods to optimise spatial-exploitation CNN models for time efficiency and classification accuracy: hyperparameter-tuning, and human-in-the-loop (HITL). We apply grid-search to optimise the hyperparameter space, whilst HITL is used to identify whether the time-to-accuracy relationship of the optimised model can be improved. To show the versatility of combining the two methods, CIFAR-10, MNIST, and Imagenette are used as model input. This paper contributes to spatial-exploitation CNN optimisation by combining hyperparameter-tuning and HITL. Results show that this combination improves classification accuracy by 1.47-2.34% and reduces the time taken to conduct this task by 27-28%, depending on dataset. We conclude that combining hyperparameter-tuning and HITL are a viable approach to optimise spatial-exploitation CNNs for devices with limited computational resources.

Keywords: deep learning; convolutional neural network; CNN; image classification; hyperparameter-tuning; human-in-the-loop; HITL.

DOI: 10.1504/IJAISC.2024.139607

International Journal of Artificial Intelligence and Soft Computing, 2024 Vol.8 No.2, pp.147 - 158

Received: 17 Dec 2022
Accepted: 04 Jan 2024
Published online: 04 Jul 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article