Consistency of Multiagent Distributed Generative Adversarial Networks.

PSGAN: Revisit the binary discriminator and an alternative for face frontalization

Optimizing the discriminator in Generative Adversarial Networks (GANs) to completion in the inner training loop is computationally prohibitive, and on finite datasets would result in overfitting. To address this, a common update strategy is to alternate between k optimization steps for the discriminator D and one optimization step for the generator G. This strategy is repeated in various GAN algorithms where k is selected empirically. In this paper, we show that this update strategy is not optimal in terms of accuracy and convergence speed, and propose a new update strategy for networks with Wasserstein GAN (WGAN) group related loss functions (e.g. WGAN, WGAN-GP, Deblur GAN, and Super resolution GAN). The proposed update strategy is based on a loss change ratio comparison of G and D. We demonstrate that the proposed strategy improves both convergence speed and accuracy.

In recent years, extensive research has been done to obtain better demodulation performance by combining signal demodulation with deep learning algorithms, such as convolutional neural networks (CNN). However, the demodulation performance of CNN is not much improved compared with traditional coherent demodulation algorithms. This paper proposes a signal demodulation algorithm based on generative adversarial networks (GAN). The proposed GAN-based demodulation network uses the confrontation characteristics between the generator and the discriminator in GAN to greatly improve the demodulation ability of the signal. Numerical results show that the demodulation performance is better than the coherent demodulation algorithm and the CNN-based demodulation algorithm.

Metasurfaces have been developed as a promising approach for manipulating electromagnetic waves. Recently, deep learning algorithms have been introduced to design metasurfaces, but the network can only output one solution for each desired input and suffers from nonunique issue. To overcome the aforementioned challenges, a deep neural network model for inverse designs of anisotropic metasurfaces with full phase properties in ultrawideband is proposed. Given the target reflection spectra as inputs, the candidate metasurface patterns are generated through a generative adversarial network (GAN), and the corresponding predictions are simply achieved by the accurate forward neural network model to match the target spectra in the whole band with high fidelity. By training the generator and discriminator in GAN in an alternating order combined with setting a threshold of discriminator loss to trigger the phase prediction, the proposed method is much more efficient and consumes less time in the training process. Numerical simulations and experimental results demonstrate that the reflection phases of the generated meta‐atoms have excellent agreements with the given targets, providing an efficient way in automatically designing metasurfaces. The most important advantage of this approach over the previous schemes is to improve the design speed significantly with very good accuracy.

Learning a good image prior is a long-term goal for image restoration and manipulation. While existing methods like deep image prior (DIP) capture low-level image statistics, there are still gaps toward an image prior that captures rich image semantics including color, spatial coherence, textures, and high-level concepts. This work presents an effective way to exploit the image prior captured by a generative adversarial network (GAN) trained on large-scale natural images. As shown in Fig. 1, the deep generative prior (DGP) provides compelling results to restore missing semantics, e.g., color, patch, resolution, of various degraded images. It also enables diverse image manipulation including random jittering, image morphing, and category transfer. Such highly flexible effects are made possible through relaxing the assumption of existing GAN-inversion methods, which tend to fix the generator. Notably, we allow the generator to be fine-tuned on-the-fly in a progressive manner regularized by feature distance obtained by the discriminator in GAN. We show that these easy-to-implement and practical changes help preserve the reconstruction to remain in the manifold of nature image, and thus lead to more precise and faithful reconstruction for real images. Code is at https://github.com/XingangPan/deep-generative-prior .

Face frontalization is one way to overcome the pose variation problem, which simplifies multi-view recognition into one canonical-view recognition. This paper presents a multi-task learning approach based on the generative adversarial network (GAN) that learns the emotion-preserving representations in the face frontalization framework. Taking advantage of adversarial relationship between the generator and the discriminator in GAN, the generator can frontalize input non-frontal face images into frontal face images while preserving the identity and expression characteristics; in the meantime, it can employ the learnt emotion-preserving representations to predict the expression class label from the input face. The proposed network is optimized by combining both synthesis and classification objective functions to make the learnt representations generative and discriminative simultaneously. Experimental results demonstrate that the proposed face frontalization system is very effective for expression recognition with large head pose variations.

De novo drug design with desired biological activitiesis crucialfor developing novel therapeutics for patients. The drug developmentprocess is time- and resource-consuming, and it has a low probabilityof success. Recent advances in machine learning and deep learningtechnology have reduced the time and cost of the discovery processand therefore, improved pharmaceutical research and development. Inthis paper, we explore the combination of two rapidly developing fieldswith lead candidate discovery in the drug development process. First,artificial intelligence has already been demonstrated to successfullyaccelerate conventional drug design approaches. Second, quantum computinghas demonstrated promising potential in different applications, suchas quantum chemistry, combinatorial optimizations, and machine learning.This article explores hybrid quantum-classical generative adversarialnetworks (GAN) for small molecule discovery. We substituted each elementof GAN with a variational quantum circuit (VQC) and demonstrated thequantum advantages in the small drug discovery. Utilizing a VQC inthe noise generator of a GAN to generate small molecules achievesbetter physicochemical properties and performance in the goal-directedbenchmark than the classical counterpart. Moreover, we demonstratethe potential of a VQC with only tens of learnable parameters in thegenerator of GAN to generate small molecules. We also demonstratethe quantum advantage of a VQC in the discriminator of GAN. In thishybrid model, the number of learnable parameters is significantlyless than the classical ones, and it can still generate valid molecules.The hybrid model with only tens of training parameters in the quantumdiscriminator outperforms the MLP-based one in terms of both generatedmolecule properties and the achieved KL divergence. However, the hybridquantum-classical GANs still face challenges in generating uniqueand valid molecules compared to their classical counterparts.

As a promising generative modeling method, Generative Adversarial Networks are a deep-learning-based generative model, in which two networks, namely the generative network and the discriminant network, contest with each other in a game during which the generator “learns” and get trained to be able to fool the discriminator of believing a specific image is (superficially) authentic. It is widely used to generate pictures based on existing data. Based on the MNIST data set, this paper studies and analyzes the application of Generative Adversarial Networks in the generation of handwritten digital images. In addition, by studying the Deep Convolutional Generative Adversarial Network and the Conditional Generation Adversarial Network model, we compared the connections and differences between these three models in training the MNIST dataset to generate handwritten digits algorithms, and gave an optimization method for generating handwritten digits.

Generative Adversarial Networks, as a powerful generative model, show great potential in generating adversarial samples and defending against adversarial attacks. In this paper, using Generative Adversarial Networks (GANs) as the basic framework, we design a network attack sample generation method based on Deep Convolutional Generative Adversarial Networks (DCGANs) and an adversarial sample defence method based on multi-scale GANs, and verify the practicality of the two methods through experiments, respectively. Compared with the three adversarial sample generation methods of AE-CDA, AE-DEEP and AE-ATTACK, the DCGAN-based adversarial sample generation method in this paper can interfere with the detection function of the anomaly detection model more effectively, and has better stability and versatility, and can maintain a relatively stable attack effect on a wide range of models and datasets. On the MNIST dataset, the classification accuracy of the adversarial sample defence method proposed in this paper is only slightly lower than that of the APE-GAN defence method on the JSMA adversarial samples, with a maximum classification accuracy of 98.69%. The maximum classification accuracy reaches 98.69%, and the time consumption is 1.5 s, which is only slightly larger than that of the APE-GAN defence method of 1.2 s. Thus, the time consumption of this paper’s multi-scale GAN-based adversarial sample defense method is smaller or equal to that of other comparative defense methods when systematic errors are ignored. The purpose of this paper is to provide a technical reference on how to eliminate adversarial perturbations using generative adversarial networks.

Deep Reinforcement Learning (DRL) has been successfully applied in several research domains such as robot navigation and automated video game playing. However, these methods require excessive computation and interaction with the environment, so enhancements on sample efficiency are required. The main reason for this requirement is that sparse and delayed rewards do not provide an effective supervision for representation learning of deep neural networks. In this study, Proximal Policy Optimization (PPO) algorithm is augmented with Generative Adversarial Networks (GANs) to increase the sample efficiency by enforcing the network to learn efficient representations without depending on sparse and delayed rewards as supervision. The results show that an increased performance can be obtained by jointly training a DRL agent with a GAN discriminator. ---- Derin Pekistirmeli Ogrenme, robot navigasyonu ve otomatiklestirilmis video oyunu oynama gibi arastirma alanlarinda basariyla uygulanmaktadir. Ancak, kullanilan yontemler ortam ile fazla miktarda etkilesim ve hesaplama gerektirmekte ve bu nedenle de ornek verimliligi yonunden iyilestirmelere ihtiyac duyulmaktadir. Bu gereksinimin en onemli nedeni, gecikmeli ve seyrek odul sinyallerinin derin yapay sinir aglarinin etkili betimlemeler ogrenebilmesi icin yeterli bir denetim saglayamamasidir. Bu calismada, Proksimal Politika Optimizasyonu algoritmasi Uretici Cekismeli Aglar (UCA) ile desteklenerek derin yapay sinir aglarinin seyrek ve gecikmeli odul sinyallerine bagimli olmaksizin etkili betimlemeler ogrenmesi tesvik edilmektedir. Elde edilen sonuclar onerilen algoritmanin ornek verimliliginde artis elde ettigini gostermektedir.

Inspired by classic Generative Adversarial Networks (GAN), we propose a novel end-to-end adversarial neural network, called SegAN, for the task of medical image segmentation. Since image segmentation requires dense, pixel-level labeling, the single scalar real/fake output of a classic GAN's discriminator may be ineffective in producing stable and sufficient gradient feedback to the networks. Instead, we use a fully convolutional neural network with new activation function in the last layer as the segmentor to generate segmentation label maps, and propose a novel adversarial critic network with a multi-scale L 1 loss function to force the critic and segmentor to learn both global and local features that capture long- and short-range spatial relationships between pixels. We show that such a SegAN framework is more effective in the segmentation task and more stable to train, and it outperforms current state-of-the-art segmentation methods in the ISBI International Skin Imaging Collaboration (ISIC) 2017 challenge, Part I Lesion Segmentation.

Federated learning (FL), as a distributed learning mechanism, can have model training completed without directly uploading original data, effectively reducing the risk of privacy leakage. However, through the shared gradient information, research shows that adversaries may reconstruct the original data. To further protect the privacy of federated learning, a federated learning defense scheme is proposed based on generative adversarial networks (GAN), which is combined with adaptive differential privacy. Firstly, the real data distribution features are learned through GAN, and replaceable pseudo data are generated. Then, the pseudo data are added with adaptive noise. Finally, the pseudo gradient generated by the pseudo data in the model is used to replace the real gradient so that adversaries cannot obtain the real gradient to further protect the privacy of user data. After simulation experiments are carried out on the MNIST dataset, the algorithm is verified using the gradient attack method. The experimental results show that the proposed algorithm is superior to the federated learning algorithm based on differential privacy in accuracy. Compared with the FedAvg algorithm, only 0.48% accuracy is lost. Therefore, it achieves a good balance between algorithm accuracy and data privacy.

SummaryFréchet inception distance (FID) has gained a better reputation as an evaluation metric for generative adversarial networks (GANs). However, it is subjected to fluctuation, namely, the same GAN model, when trained at different times can have different FID scores, due to the randomness of the weight matrices in the networks, stochastic gradient descent, and the embedded distribution (activation outputs at a hidden layer). In calculating the FIDs, embedded distribution plays the key role and it is not a trivial question from where obtaining it since it contributes to the fluctuation also. In this article, I showed that embedded distribution can be obtained from three different subspaces of the weight matrix, namely, from the row space, the null space, and the column space, and I analyzed the effect of the each space to Fréchet distances (FDs). Since the different spaces show different behaviors, choosing a subspace is not an insignificant decision. Instead of directly using the embedded distribution obtained from hidden layer's activations to calculate the FD, I proposed to use projection of embedded distribution onto the null space of the weight matrix among the three subspaces to avoid the fluctuations. My simulation results conducted at MNIST, CIFAR10, and CelebA datasets, show that, by projecting the embedded distributions onto the null spaces, possible parasitic effects coming from the randomness are being eliminated and reduces the number of needed simulationsin MNIST dataset,in CIFAR10, andin CelebA dataset.

In the last few years, generative adversarial networks(GANs) based on the concept of game theory have become one of the key research in the field of image generation. Nevertheless, the CNN network architecture in GANs cannot effectively utilize spatial information, which is meaningful for the classification problem. The capsule network proposed in 2017 can solve the problems of CNN to a certain extent, which replaces individual neurons in a neural network with a vector of neurons that are wrapped together and form a capsule and can encode several feature attributes and spatial information of an image. Consequently, the capsule network is combined with GANs, the least squares loss is used to optimize and strengthen the network, and the LS-CapsGAN image generation model is proposed, which makes the training safer and the generated image quality higher. In this paper, the advantages of LS-CapsGAN in image data distribution modeling on MNIST, Fashion-MNIST, and CIFAR-10 datasets are expounded through three groups of comparative experiments, which verify that LS-CapsGAN has a certain potential in the field of image generation.

As a critical and prevalent service in future mobile networks, virtual reality (VR) is latency-sensitive and power-hungry, bringing out the optimization problem of trade-off among power saving, delay, and resource utilization. Content caching and render offloading are deemed as promising solutions to meet the stringent requirements of VR on data transmission speed and end-to-end latency. In this article, we propose a novel distributed computing framework based on multi-agent deep deterministic policy gradient (MADDPG) for joint optimizing terminal-cooperative caching and offloading for VR tasks. Since the individual VR user can hardly reach the optimal actions based on its limited local observed states and samples, MADDPG with centralized training and distributed execution is exploited to solve the above challenge. In addition, the generative adversarial network (GAN) is introduced to obtain experience-enhanced agents in the offline training phase and to achieve an optimal allocation to minimize energy consumption in the online inferring phase. The Nash equilibrium is proven in the case that the distribution of finite real VR data samples is well imitated and complemented by GAN. Numerical results demonstrate that our algorithm has significant superiorities in terms of convergence performance and energy consumption over other benchmarks.