MUMBAI, India, Feb. 13 -- Intellectual Property India has published a patent application (202641007653 A) filed by Gorla Venkata Ganesh; and Koneru Lakshmaiah Education Foundation, Vijayawada, Andhra Pradesh, on Jan. 27, for 'a metaheuristic-based system and method for hyperparameter optimization of convolutional neural networks using whale optimization algorithm variants.'
Inventor(s) include G. Radha Bai; Dr. Sateesh Kumar Deevi; and Gorla Venkata Ganesh.
The application for the patent was published on Feb. 13, under issue no. 07/2026.
According to the abstract released by the Intellectual Property India: "A Metaheuristic-Based System and Method for Hyperparameter Optimization of Convolutional Neural Networks Using Whale Optimization Algorithm Variants Hyperparameter optimization (HPO) plays a critical role in improving the performance of Convolutional Neural Networks (CNNs). Existing reviews have focused on algorithms such as Bayesian Optimization, Tree-Structured Parzen Estimator, and metaheuristics like Genetic Algorithm, Particle Swarm Optimization, Ant Colony Optimization, Firefly Algorithm, Gray Wolf Optimization, and Harmony Search. However, Whale Optimization Algorithm (WOA), a nature-inspired metaheuristic based on humpback whales' bubble-net hunting strategy, has emerged as a promising alternative for CNN tuning. This extended review synthesizes the original taxonomy of ten HPO algorithms and introduces WOA and its variants (IWOA, CWOA, EWOA) into the comparative framework. Reported results show WOA achieving up to 99.63% accuracy on MNIST and competitive performance on Fashion-MNIST, alongside strong outcomes in non-vision tasks such as forecasting and segmentation. We provide comparative tables, practical guidelines, and highlight WOA's strengths and limitations relative to other methods. This work aims to guide researchers in selecting suitable HPO strategies for CNNs under varying computational budgets and application domains."
Disclaimer: Curated by HT Syndication.
