A comprehensive review of generative adversarial networks: Fundamentals, applications, and challenges

Abstract

In machine learning, a generative model is responsible for generating new samples of data in terms of a probabilistic model. Generative adversarial network (GAN) has been widely used to generate realistic samples in different domains and outperforms its peers in the generative models family. However, producing a robust GAN model is not a trivial task because many challenges face the GAN during the training process and impact its performance, affecting the quality and diversity of the generated samples. In this article, we conduct a comprehensive review of GANs to present the fundamentals of GAN, including its components, types, and objective functions. Also, we present an overview of the evaluation matrices used to evaluate GAN models. Moreover, we list the applications of GANs and research work in various domains. Finally, we present the challenges that face GANs and highlight two significant issues, representing mode collapse and training instability, in addition to those research efforts that tackle these challenges.This article is categorized under: Statistical Learning and Exploratory Methods of the Data Sciences > Deep Learning Statistical Learning and Exploratory Methods of the Data Sciences > Neural Networks

MRD‐GAN: Multi‐representation discrimination GAN for enhancing the diversity of the generated data

“…In the context of evaluating GAN performance, there are several metrics used to evaluate the distorted images from several perspectives such as image quality and diversity. 5,44 According to Reference 44, the Fréchet Inception Distance (FID) is sensitive metric to mode dropping and more plausible than other metrics. FID quantifies the distance between real images and generated images at the feature level based on the pre-trained Inception-V3 model, but with customized calculations in the final pooling layer.…”

“…1,2 It is worth noting that the generative adversarial network (GAN) has recently made significant strides and has outperformed its peers in the generative models family in various tasks, such as enhancing image resolution, completing damaged parts of images, translating data between different domains, predicting new human behaviors and expected poses, generating synthetic data for a specific task, improving speech, and removing noisy signals. 1,[3][4][5] A GAN is a generative model that learns the data distribution in an unsupervised manner. As illustrated in Figure 1, the fundamental form of GAN comprises two networks, the generator and discriminator.…”