Fictitious GAN: Training GANs with Historical Models

Ge, Hao; Yin, Xia; Chen, Xu; Berry, Randall A.; Wu, Ying

doi:10.1007/978-3-030-01246-5_8

Cited by 18 publications

(15 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The generated images by the generator network are then classified as real or fake by the discriminator network during the training stage, i.e., D: y [0,1]. This two-player zero-sum game continues until satisfactory image generation is achieved [29]. Training of GAN algorithms is subject to research, and it is a continuing effort to handle efficient training of GAN models [30].…”

Section: Conditional Generative Adversarial Network (Cgan)mentioning

confidence: 99%

A comparison of deep learning algorithms on image data for detecting floodwater on roadways

et al. 2022

View full text Add to dashboard Cite

Object detection and segmentation algorithms evolved significantly in the last decade. Simultaneous object detection and segmentation paved the way for real-time applications such as autonomous driving. Detection and segmentation of (partially) flooded roadways are essential inputs for vehicle routing and traffic management systems. This paper proposes an automatic floodwater detection and segmentation method utilizing the Mask Region-Based Convolutional Neural Networks (Mask-R-CNN) and Generative Adversarial Networks (GAN) algorithms. To train the model, manually labeled images with urban, suburban, and natural settings are used. The performances of the algorithms are assessed in accurately detecting the floodwater captured in images. The results show that the proposed Mask-R-CNN-based floodwater detection and segmentation outperform previous studies, whereas the GAN-based model has a straightforward implementation compared to other models.

show abstract

Section: Conditional Generative Adversarial Network (Cgan)mentioning

confidence: 99%

A comparison of deep learning algorithms on image data for detecting floodwater on roadways

et al. 2022

View full text Add to dashboard Cite

show abstract

“…In recent years, a lot of research has been targeted at solving this problem and one of them has indicated that if, at the point where equilibrium is achieved, the eigenvalues of the Jacobian only falls into the negative realpart, the training of GAN can converge locally with a small learning rate [27], [28]. It has been proved that the optimal solution to (1) with p data (x) = q G (x) and D * (x) = 0.5 is actually a unique Nash equilibrium of the game [15]. So theoretically, making sure that the discriminator converging to 0.5 is key to finding the Nash equilibrium in GANs.…”

Section: B Nash Equilibriummentioning

confidence: 99%

Constrained Generative Adversarial Networks

Chao

Cao

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Generative Adversarial Networks (GANs) are a powerful subclass of generative models. Yet, how to effectively train them to reach Nash equilibrium is a challenge. A number of experiments have indicated that one possible solution is to bound the function space of the discriminator. In practice, when optimizing the standard loss function without limiting the discriminator's output, the discriminator may suffer from lack of convergence. To be able to reach the Nash equilibrium in a faster way during training and obtain better generative data, we propose constrained generative adversarial networks, GAN-C, where a constraint on the discriminator's output is introduced. We theoretically prove that our proposed loss function shares the same Nash equilibrium as the standard one, and our experiments on mixture of Gaussians, MNIST, CIFAR-10, STL-10, FFHQ, and CAT datasets show that our loss function can better stabilize training and yield even better high-quality images.

show abstract

“…Although the authors in [8] presented a few game-model GANs, they have not done a comprehensive survey in this field, and many new pieces of research have not been covered. We hope that our survey will serve as a reference for interested Proposed GAN Taxonomy Modified Game Model Stochastic game [45] Stackelberg game [46], [47] Bi-affine game [48] Modified Learning Method No regret learning [10], [49], [50] Fictitious play [27] Federated learning [51], [52] Reinforcement learning [4], [53]- [63] Modified Architecture Multiple generators, One discriminator [46], [64]- [67] One generator, Multiple discriminators [60], [68]- [72] Multiple generators, Multiple discriminators [51], [66], [73] One generator, One discriminator, One classifier [4], [74] One generator, One discriminator, One RL agent [58], [59], [75], [76] Fig. 2: The proposed taxonomy of the GAN advances by game theory.…”

Section: Gan Applicationsmentioning

confidence: 99%

“…By relating GAN with the two-player zero-sum game, Ge et al in [27] design a training algorithm to simulate the fictitious play on GAN and provide a theoretical convergence guarantee. They also show that by assuming the best response at each update in Fictitious GAN, the distribution of the mixture outputs from the generators converges to the data distribution.…”

Section: Md-ganmentioning

confidence: 99%

“…The discriminator outputs converge to the optimum discriminator function. The authors in [27] use two queues D and G, to store the historically trained models of the discriminator and the generator. They also show that Fictitious GAN can effectively resolve some convergence issues that the standard training approach cannot resolve and can be applied on top of existing GAN variants.…”

Section: Md-ganmentioning

confidence: 99%

See 1 more Smart Citation

Game of GANs: Game-Theoretical Models for Generative Adversarial Networks

Moghadam¹,

Boroomand²,

Barzegar-Jalali³

et al. 2021

Preprint

View full text Add to dashboard Cite

Generative Adversarial Network, as a promising research direction in the AI community, recently attracts considerable attention due to its ability to generating high-quality realistic data. GANs are a competing game between two neural networks trained in an adversarial manner to reach a Nash equilibrium. Despite the improvement accomplished in GANs in the last years, there remain several issues to solve. In this way, how to tackle these issues and make advances leads to rising research interests. This paper reviews literature that leverages the game theory in GANs and addresses how game models can relieve specific generative models' challenges and improve the GAN's performance. In particular, we firstly review some preliminaries, including the basic GAN model and some game theory backgrounds. After that, we present our taxonomy to summarize the state-of-the-art solutions into three significant categories: modified game model, modified architecture, and modified learning method. The classification is based on the modifications made in the basic model by the proposed approaches from the game-theoretic perspective. We further classify each category into several subcategories. Following the proposed taxonomy, we explore the main objective of each class and review the recent work in each group. Finally, we discuss the remaining challenges in this field and present the potential future research topics.

show abstract

Fictitious GAN: Training GANs with Historical Models

Cited by 18 publications

References 15 publications

A comparison of deep learning algorithms on image data for detecting floodwater on roadways

A comparison of deep learning algorithms on image data for detecting floodwater on roadways

Constrained Generative Adversarial Networks

Game of GANs: Game-Theoretical Models for Generative Adversarial Networks

Contact Info

Product

Resources

About