Modified symbiotic organisms search optimization for automatic construction of convolutional neural network architectures

• Published in: Intelligent systems with applications 2024-06, Vol.22, p.200349, Article 200349
• Main Authors: Jauro, Fatsuma, Ya'u Gital, Abdulsalam, Abdullahi, Usman Ali, Abdulsalami, Aminu Onimisi, Abdullahi, Mohammed, Ibrahim, Adamu Abubakar, Chiroma, Haruna
• Format: Article
• Language: English
• Subjects: And deep learning; Convolutional neural network; Neural architecture search; Symbiotic organism search
• ISSN: 2667-3053; 2667-3053
• DOI: 10.1016/j.iswa.2024.200349

•An algorithm that automatically generates convnet architectures using the symbiotic organism search optimization algorithm has been proposed.•A novel mutual vecto operator is introduced to enable average computation between two organisms with different architectural structures.•A novel sum operator that applies a simple addition rule to compute the sum of two individual architectures that might differ in terms of the number of layers and parameters.•A novel parasite vector generation strategy is introduced, which randomly selects one among multiple mutation strategies to obtain the parasite individual. Convolutional Neural Networks (ConvNets) have demonstrated impressive capabilities in image classification; however, the manual creation of these models is a labor-intensive and time-consuming endeavor due to their inherent complexity. This research introduces an innovative approach to Convolutional Neural Network (ConvNet) architecture generation through the utilization of the Symbiotic Organism Search ConvNet (SOS_ConvNet) algorithm. Leveraging the Symbiotic Organism Search optimization technique, SOS_ConvNet evolves ConvNet architectures tailored for diverse image classification tasks. The algorithm's distinctive feature lies in its ability to perform non-numeric computations, rendering it adaptable to intricate deep learning problems. To assess the effectiveness of SOS_ConvNet, experiments were conducted on diverse datasets, including MNIST, Fashion-MNIST, CIFAR-10, and the Breast Cancer dataset. Comparative analysis against existing models showcased the superior performance of SOS_ConvNet in terms of accuracy, error rate, and parameter efficiency. Notably, on the MNIST dataset, SOS_ConvNet achieved an impressive 0.31 % error rate, while on Fashion-MNIST, it demonstrated a competitive 6.7 % error rate, coupled with unparalleled parameter efficiency of 0.24 million parameters. The model excelled on CIFAR-10 and BreakHis datasets, yielding accuracies of 82.78 % and 89.12 %, respectively. Remarkably, the algorithm achieves remarkable accuracy while maintaining moderate model size.