Improving Semi-Supervised Learning Generative Adversarial Networks
| dc.contributor.author | Moolla, Faheem | |
| dc.contributor.co-supervisor | Bau, Hairong | |
| dc.contributor.supervisor | Van Zyl, Terence | |
| dc.date.accessioned | 2024-10-18T12:33:17Z | |
| dc.date.available | 2024-10-18T12:33:17Z | |
| dc.date.issued | 2023-08 | |
| dc.description | A research report submitted in partial fulfillment of the requirements for the degree of MSc (CW/RR) in Artificial Intelligence, to the Faculty of Science, in the School of Computer Science and Applied Mathematics, University of the Witwatersrand, Johannesburg, 2023. | |
| dc.description.abstract | Generative Adversarial Networks (GANs) have shown remarkable potential in generating high-quality images, with semi-supervised GANs providing a high classification accuracy. In this study, an enhanced semi supervised GAN model is proposed wherein the generator of the GAN is replaced by a pre-trained decoder from a Variational Autoencoder. The model presented outperforms regular GAN and semi-supervised GAN models during the early stages of training, as it produces higher quality images. Our model demonstrated significant improvements in image quality across three datasets - namely the MNIST, Fashion MNIST, and CIFAR-10 datasets - as evidenced by higher accuracies obtained from a Convolutional Neural Network (CNN) trained on generated images, as well as superior inception scores. Additionally, our model prevented mode collapse and exhibited smaller oscillations in the discriminator and generator loss graphs compared to baseline models. The presented model also provided remarkably high levels of classification accuracy, by obtaining 99.32% on the MNIST dataset, 92.78% on the Fashion MNIST dataset, and 83.22% on the CIFAR-10 dataset. These scores are notably robust as they improved some of the classification accuracies obtained by two state-of-the-art models, indicating that the presented model is a significantly improved semi-supervised GAN model. However, despite the high classification accuracy for the CIFAR-10 dataset, a considerable drop in accuracy was observed when comparing generated images to real images for this dataset. This suggests that the quality of those generated images can be bettered and the presented model performs better with less complex datasets. Future work could explore techniques to enhance our model’s performance with more intricate datasets, ultimately expanding its applicability across various domains. | |
| dc.description.submitter | MM2024 | |
| dc.faculty | Faculty of Science | |
| dc.identifier | 0000-0002-4875-9089 | |
| dc.identifier.citation | Moolla, Faheem. (2023). Improving Semi-Supervised Learning Generative Adversarial Networks. [Master's dissertation, University of the Witwatersrand, Johannesburg]. https://hdl.handle.net/10539/41694 | |
| dc.identifier.uri | https://hdl.handle.net/10539/41694 | |
| dc.language.iso | en | |
| dc.publisher | University of the Witwatersrand, Johannesburg | |
| dc.rights | ©2023 University of the Witwatersrand, Johannesburg. All rights reserved. The copyright in this work vests in the University of the Witwatersrand, Johannesburg. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of University of the Witwatersrand, Johannesburg. | |
| dc.rights.holder | University of the Witwatersrand, Johannesburg | |
| dc.school | School of Computer Science and Applied Mathematics | |
| dc.subject | Generative Adversarial Networks (GANs) | |
| dc.subject | Variational Autoencoder | |
| dc.subject | onvolutional Neural Network (CNN) | |
| dc.subject | MNIST | |
| dc.subject | Fashion MNIST | |
| dc.subject | CIFAR-10 | |
| dc.subject | UCTD | |
| dc.subject.other | SDG-9: Industry, innovation and infrastructure | |
| dc.title | Improving Semi-Supervised Learning Generative Adversarial Networks | |
| dc.type | Dissertation |