Enhancing face recognition accuracy in authentication systems through hyperparameter tuning of deep learning models
Shahina Anwarul · 2025
In the realm of face recognition, the accuracy of deep learning models is paramount. This chapter delves into the critical role of hyperparameter tuning in enhancing the performance of these models. By systematically adjusting hyperparameters, we can significantly improve model accuracy, leading to more reliable and efficient face recognition systems. We explore various hyperparameters, such as batch size, learning rate, epoch count, data augmentation, and network architecture, and demonstrate their impact through a series of experiments using samples of Labelled Faces in the Wild (LFW) dataset. The chapter provides a comprehensive guide on the methodologies for hyperparameter tuning and presents empirical results that highlight its effectiveness. Our findings underscore the importance of meticulous hyperparameter optimization in developing state-of-the-art face recognition models, ultimately contributing to advancements in security, authentication, and beyond.