Quantum Inspired Grey Wolf Optimizer for Convolutional Neural Network Hyperparameter Optimization

Convolutional Neural Networks (CNNs) have revolutionized the field of computer vision and achieved remarkable success in various image-related tasks. However, their performance heavily depends on the correct tuning of hyperparameters, which can be challenging and time-consuming due to the vast search space. Recent advances in metaheuristic optimization algorithms have demonstrated their effectiveness in estimating the hyperparameters of deep learning networks. On the other hand, introducing Quantum Computing (QC) concepts into metaheuristic algorithms has further propelled this research field. This paper proposes a novel approach, qGWO-CNN, to fine-tune CNN hyperparameters. Our proposed qGWO-CNN method uses an improved variant of the Grey Wolf Optimizer (GWO), inspired by QC principles, to search for optimal hyperparameter values efficiently. To evaluate qGWO-CNN performance, we conduct experiments on the CIFAR-10 dataset using the AlexNet architecture. Comparison with other GWO-based approaches shows the excellent performance of the proposed algorithm in terms of classification accuracy. Furthermore, the numerical results indicate the potential of qGWO-CNN to be a promising approach for improving CNN performance.