A Survey of Generative Adversarial Networks

Abstract

Generative Adversarial Networks (GANs) are a thriving unsupervised machine learning technique that has led to significant advances in various fields such as computer vision, natural language processing, among others. However, GANs are known to be difficult to train and usually suffer from mode collapse and the discriminator winning problem. To interpret the empirical observations of GANs and design better ones, we deconstruct the study of GANs into three components and make the following contributions.• Formulation: we propose a perturbation view of the population target of GANs. Building on this interpretation, we show that GANs can be connected to the robust statistics framework, and propose a novel GAN architecture, termed as Cascade GANs, to provably recover meaningful low-dimensional generator approximations when the real distribution is high-dimensional and corrupted by outliers. • Generalization: given a population target of GANs, we design a systematic principle, projection under admissible distance, to design GANs to meet the population requirement using only finite samples. We implement our principle in three cases to achieve polynomial and sometimes near-optimal sample complexities: (1) learning an arbitrary generator under an arbitrary pseudonorm;(2) learning a Gaussian location family under total variation distance, where we utilize our principle to provide a new proof for the nearoptimality of the Tukey median viewed as GANs;(3) learning a low-dimensional Gaussian approximation of a high-dimensional arbitrary distribution under Wasserstein distance. We demonstrate a fundamental trade-off in the approximation error and statistical error in GANs, and demonstrate how to apply our principle in practice with only empirical samples to predict how many samples would be sufficient for GANs in order not to suffer from the discriminator winning problem. • Optimization: we demonstrate alternating gradient descent is provably not locally asymptotically stable in optimizing the GAN formulation of PCA. We diagnose the problem as the minimax duality gap being non-zero, and propose a new GAN architecture whose duality gap is zero, where the value of the game is equal to the previous minimax value (not the maximin value). We prove the new GAN architecture is globally asymptotically stable in solving PCA under alternating gradient descent.

Accuracy of Using Generative Adversarial Networks for Glaucoma Detection: Systematic Review and Bibliometric Analysis

“…Among these categories, it is notable in Figure 5 that the first 4 layers classified articles based on the method used (direct, hierarchical, or iterative) [ 32 ], the model structure [ 65 ], the architecture category (optimization function or structure and conditional based) [ 66 - 68 ], and the generator’s backbone (CNN based or U-Net based) [ 69 ] consecutively.…”

“…The subsequent layer in our taxonomy was to classify methods according to the generator’s backbone (eg, U-Net based or CNN based) [ 69 ]. Papers [ 42 , 46 , 49 , 51 , 57 , 73 , 75 , 76 , 80 - 87 ] represented about 50% of the studies (n=16) and were U-Net-based architectures.…”

“…GAN is an inclusive system that can be combined with various deep learning models to address problems that conventional ML algorithms cannot solve, such as poor quality of outputs, insufficient training samples, and deep feature extraction [ 68 ]. Furthermore, it outperforms conventional methods in editing and synthesizing image [ 69 ]. GAN allied with transfer learning can effectively reduce misjudgment of OD/OC in glaucoma cases and improve accuracy and generalization capability; however, better backbone network and different upsampling methods are required to improve performance [ 86 ] and exploring other downstream tasks may enhance the model’s performance [ 93 ].…”

“…However, in our taxonomy, we combined them all. Furthermore, we added 4 more classification criteria as follows: (1) method architecture (direct, hierarchical, iterative) [ 32 ]; (2) model structure (2 players, multiple players) [ 65 ]; (3) GAN category (optimization function, structure, and conditional) [ 66 - 68 ]; and (4) generator backbone (U-Net based or CNN based) [ 69 ].…”

Deep CNN and Deep GAN in Computational Visual Perception‐Driven Image Analysis

“…Tensor is the input, which is a multidimensional array, the kernel is a small array of numbers, and the feature map is the output tensor, shown in Figure 3. e convolution operation is a linear process where a dot product is performed between the A survey of generative adversarial networks [33] State-of-the-art GAN architectures are surveyed, and their application domains on natural language processing and computer vision are discussed. e loss functions of the GAN variants are discussed.…”