Troika Generative Adversarial Network (T-GAN): A Synthetic Image Generator That Improves Neural Network Training for Handwriting Classification

• Main Authors: Milan, Joe Anthony M, Fernandez, Proceso L, Jr
• Format: text
• Published: Archīum Ateneo 2020
• Subjects: artificial neural networks (ANN); data augmentation; generative adversarial network (GAN); handwriting classification; synthetic data; troika GAN; Computer Sciences
• Online Access: https://archium.ateneo.edu/discs-faculty-pubs/234; https://archium.ateneo.edu/cgi/viewcontent.cgi?article=1230&context=discs-faculty-pubs
• Institution: Ateneo De Manila University

Training an artificial neural network for handwriting classification requires a sufficiently sized annotated dataset in order to avoid overfitting. In the absence of sufficient instances, data augmentation techniques are normally considered. In this paper, we propose the troika generative adversaria...