Authors : A. W. C. K. Atugoda; S. D. Fernando

Volume/Issue : Volume 10 - 2025, Issue 11 - November

Google Scholar : https://tinyurl.com/mw96uwm4

Scribd : https://tinyurl.com/yujf2r2z

DOI : https://doi.org/10.38124/ijisrt/25nov358

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Note : Google Scholar may take 30 to 40 days to display the article.

Abstract : Deep Neural Networks (DNNs) such as ResNet-50 have achieved state-of-the-art results in large-scale image classification. However, their training performance depends strongly on optimization efficiency. Conventional methods like Stochastic Gradient Descent (SGD) often struggle with slow convergence, learning-rate sensitivity, and local minima entrapment. To overcome these challenges, this study introduces an Enhanced Particle Swarm Optimization (PSO) technique that adaptively adjusts particle dynamics to maintain a better exploration-exploitation balance during training. The method was evaluated on the CIFAR-10 dataset and compared against Standard PSO and SGD under identical experimental conditions. The results demonstrate that Enhanced PSO achieves superior validation accuracy (up to 95.7%), faster convergence, and a more stable weight distribution centered near zero. These characteristics reflect a well-regularized learning process with improved generalization. Overall, the Enhanced PSO framework provides a robust and scalable optimization approach for deep neural networks, offering a viable alternative to conventional gradient-based training algorithms.

Keywords : Particle Swarm Optimization, Deep Neural Networks, ResNet-50, Convergence Stability, Weight Values Optimization.

References :

1. Zhang H, et al. Developing a novel artificial intelligence model to estimate the capital cost of mining projects using deep neural network-based ant colony optimization algorithm. Resour Policy. 2020;66:101604. doi:10.1016/j.resourpol.2020.101604.
2. Aje OF, Josephat AA. Global Journal of Engineering and Technology Advances. Glob J Eng Technol Adv. 2020;3(3):1-6. doi:10.30574/gjeta.
3. Atugoda AWK, Fernando S. Improved Particle Swarm Optimization for Optimizing the Deep Convolutional Neural Network. In: Proceedings of the International Conference on Information Technology Research (ICITR); 2023 Dec 8–9; Moratuwa, Sri Lanka. IEEE; 2023. p. 45–50.
4. Zhao X, Li J, Liu Y, Wang C. An efficient swarm intelligence approach to the optimization on high-dimensional problems. Front Comput Neurosci. 2024;18:1283974. doi:10.3389/fncom.2024.1283974.
5. Zhang Y, Wang S, Ji G. Swarm intelligence and its applications. Sci World J. 2013;2013:528069. doi:10.1155/2013/528069.
6. Nickabadi A, Ebadzadeh MM, Safabakhsh R. A novel particle swarm optimization algorithm with adaptive inertia weight. Appl Soft Comput J. 2011;11(4):3658-70. doi:10.1016/j.asoc.2011.01.037.
7. Kennedy J, Eberhart R. Particle swarm optimization. Proc IEEE Int Conf Neural Netw. 1995;4:1942-8. doi:10.1109/ICNN.1995.488968.
8. Gad AG. Particle swarm optimization algorithm and its applications: A systematic review. Arch Comput Methods Eng. 2022;29(5):2531-61. doi:10.1007/s11831-021-09588-4.
9. Banks A, Vincent J, Anyakoha C. A review of particle swarm optimization. Part I: Background and development. Nat Comput. 2007;6(4):467-84.2. LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015;521(7553):436–444.
10. Yang XS. Nature-Inspired Optimization Algorithms. Oxford: Elsevier; 2014.
11. Serizawa T, Fujita H. Optimization of convolutional neural network using the linearly decreasing weight particle swarm optimization. arXiv preprint arXiv:2001.05670. 2020.