AuthorGAN: Improving GAN Reproducibility using a Modular GAN Framework
This addresses the problem of high skill barriers for practitioners in using GANs, offering a tool for democratization, though it is incremental as it builds on existing GAN methods.
The paper tackles the difficulty of implementing and using GAN models by proposing AuthorGAN, a modular framework that enables interoperability across libraries and a visual designer for code-free architecture design, demonstrating performance on MNIST with five GAN models.
Generative models are becoming increasingly popular in the literature, with Generative Adversarial Networks (GAN) being the most successful variant, yet. With this increasing demand and popularity, it is becoming equally difficult and challenging to implement and consume GAN models. A qualitative user survey conducted across 47 practitioners show that expert level skill is required to use GAN model for a given task, despite the presence of various open source libraries. In this research, we propose a novel system called AuthorGAN, aiming to achieve true democratization of GAN authoring. A highly modularized library agnostic representation of GAN model is defined to enable interoperability of GAN architecture across different libraries such as Keras, Tensorflow, and PyTorch. An intuitive drag-and-drop based visual designer is built using node-red platform to enable custom architecture designing without the need for writing any code. Five different GAN models are implemented as a part of this framework and the performance of the different GAN models are shown using the benchmark MNIST dataset.