How Transfer Learning Is Used in Generative Models forImage Classification: Improved Accuracy

Abstract

Recent breakthroughs in generative neural networks have paved the way for transformative capabilities,particularly in their capacity to generate novel data, notably in the realm of images. Theintegration of these models with the increasingly popular technique of transfer learning, designed forproficient feature extraction, holds the promise of enhancing overall performance. This paper delvesinto the exploration of employing generative models in conjunction with transfer learning methods forfeature extraction, with a specific focus on image classification tasks. Our investigation aims to scrutinizethe effectiveness of leveraging generative models alongside pre-trained models as feature extractorsin the context of image classification. To the best of our knowledge, our investigation is the first tolink transfer learning and generative models for a discriminative task under one roof. The proposedapproach undergoes rigorous evaluation on two distinct datasets, employing specific metrics to gaugethe model’s performance. The results exhibit a notable nearly 10% enhancement achieved through theintegration of generative models, underscoring their potential for achieving heightened accuracy inimage classification. These findings highlight significant advancements in image classification accuracy,surpassing the performance of conventional Artificial Neural Network (ANN) models.