The IRMA Community
Newsletters
Research IRM
Click a keyword to search titles using our InfoSci-OnDemand powered search:
|
Optimizing Neural Networks With Ant Colony Intelligence: A Bio-Inspired Approach to Deep Learning Architecture and Hyperparameter Tuning
|
|
Author(s): M. Robinson Joel (KCG College of Technology, Chennai, India), V. Ebenezer (Karunya Institute of Technology and Sciences, India), J. Immanuel Johnraja (Karunya Institute of Technology and Sciences, India), P. Getzi Jeba Lillipushpam (Karunya Institute of Technology and Sciences, India), M. Vargheese (PSN College of Engineering and Technology, India)and Belfin Robinson (University of North Carolina, USA)
Copyright: 2026
Pages: 26
Source title:
Metaheuristic Algorithms and Optimizing Neural Networks for Biomedical Image Processing
Source Author(s)/Editor(s): Prasanalakshmi Balaji (King Khalid University, Saudi Arabia), K. Martin Sagayam (Karunya Institute of Technology and Sciences, India), Aditi Sharma (Symbiosis International University, India)and Korhen Cengiz (University of Fujairah, UAE)
DOI: 10.4018/979-8-3373-0523-3.ch005
Purchase
|
Abstract
An effective bio-inspired method for resolving challenging optimisation problems is Ant Colony Optimisation (ACO). With an emphasis on network topology, layer configurations, and parameter tweaking, this paper investigates the use of ACO in neural network design optimisation. Neural networks may be dynamically constructed to attain higher accuracy, shorter training times, and cheaper computing costs by using the ACO's ability to traverse search regions through collective learning and pheromone-based pathfinding. We look into how ACO can be integrated with deep neural networks and how it affects architectural design and hyperparameter selection. Our findings show that, especially for workloads involving large-scale data processing, ACO-driven designs provide a notable performance benefit over conventional techniques. By offering a fresh approach to neural network optimisation, this work advances the area of artificial intelligence and creates opportunities for future investigations into bio-inspired methods in deep learning architecture design.
Related Content
|
Arshiya Begum, Asfia Sabahath.
© 2026.
36 pages.
|
|
Farica Qureshi, Satyam Sharma, Rafiya Nazir.
© 2026.
30 pages.
|
|
Inderdeep Kaur, Aleem Ali.
© 2026.
34 pages.
|
|
Sridevi Tharanidharan, Prasanalakshmi Balaji, Gabriel Xiao-Guang Yue, Renuka Devi.
© 2026.
26 pages.
|
|
M. Robinson Joel, V. Ebenezer, J. Immanuel Johnraja, P. Getzi Jeba Lillipushpam, M. Vargheese, Belfin Robinson.
© 2026.
26 pages.
|
|
V. Padmajothi, T. S. Poornappriya, C. Anuradha, S. Vijayalakshmi, R. Balasubramani, S. Harihara Gopalan.
© 2026.
18 pages.
|
|
Manoj Nagappan, Sriraman Ramalingam.
© 2026.
26 pages.
|
|
|