Generative Network Research Articles

A novel generative corrective network structure for traffic forecasting

A novel generative corrective network structure for traffic forecasting

An exploratory study of the relationship between pulse transit time and blood pressure based on causal inference

Aim: We propose to examine the causal relationship between the noninvasive features represented by pulse transit time (PTT) and blood pressure (BP), with the aim of mitigating the impact of confounding factor(s) and thus improving the performance of BP estimation. Methods: We identified the causal graph of BP and the important noninvasive features extracted from electrocardiogram (ECG) and photoplethysmogram (PPG) via fast causal inference (FCI) algorithm, with the orientations of the edges in the causal graph being determined by the causal generative neural networks (CGNN) algorithm. With the knowledge obtained from the causal graph, we further used hierarchical regression model to estimate BP, and validated the proposed method on 17 subjects. Results: We found that the obtained causal graph was almost consistent with the prior knowledge, and heart rate (HR) was one of the main confounding factors of PTT and BP. Incorporating HR into the hierarchical regression model to eliminate its confounding effect on the PTT-based BP estimation, the mean value of SBP and DBP estimation was improved by 1.27 and 1.89 mmHg, respectively, and the mean absolute difference (MAD) was improved by 2.28 and 3.60 mmHg, respectively. Conclusion: Causal inference-based method has the potential to clarify the causal relationship between BP and related wearable noninvasive features, which can further shed light on developing new methods for cuffless BP with acceptable accuracy.

Semi-TSGAN: Semi-Supervised Learning for Highlight Removal Based on Teacher-Student Generative Adversarial Network.

Despite recent notable advancements in highlight image restoration techniques, the dearth of annotated data and the lightweight deployment of highlight removal networks pose significant impediments to further advancements in the field. In this paper, to the best of our knowledge, we first propose a semi-supervised learning paradigm for highlight removal, merging the fusion version of a teacher-student model and a generative adversarial network, featuring a lightweight network architecture. Initially, we establish a dependable repository to house optimal predictions as pseudo ground truth through empirical analyses guided by the most reliable No-Reference Image Quality Assessment (NR-IQA) method. This method serves to assess rigorously the quality of model predictions. Subsequently, addressing concerns regarding confirmation bias, we integrate contrastive regularization into the framework to curtail the risk of overfitting on inaccurate labels. Finally, we introduce a comprehensive feature aggregation module and an extensive attention mechanism within the generative network, considering a balance between network performance and computational efficiency. Our experimental evaluations encompass comprehensive assessments on both full-reference and non-reference highlight benchmarks. The results demonstrate conclusively the substantive quantitative and qualitative enhancements achieved by our proposed algorithm in comparison to state-of-the-art methodologies.

Transfer-Learning-Enhanced Regression Generative Adversarial Networks for Optimal eVTOL Takeoff Trajectory Prediction

Electric vertical takeoff and landing (eVTOL) aircraft represent a crucial aviation technology to transform future transportation systems. The unique characteristics of eVTOL aircraft include reduced noise, low pollutant emission, efficient operating cost, and flexible maneuverability, which in the meantime pose critical challenges to advanced power retention techniques. Thus, optimal takeoff trajectory design is essential due to immense power demands during eVTOL takeoffs. Conventional design optimizations, however, adopt high-fidelity simulation models in an iterative manner resulting in a computationally intensive mechanism. In this work, we implement a surrogate-enabled inverse mapping optimization architecture, i.e., directly predicting optimal designs from design requirements (including flight conditions and design constraints). A trained inverse mapping surrogate performs real-time optimal eVTOL takeoff trajectory predictions with no need for running optimizations; however, one training sample requires one design optimization in this inverse mapping setup. The excessive training cost of inverse mapping and the characteristics of optimal eVTOL takeoff trajectories necessitate the development of the regression generative adversarial network (regGAN) surrogate. We propose to further enhance regGAN predictive performance through the transfer learning (TL) technique, creating a scheme termed regGAN-TL. In particular, the proposed regGAN-TL scheme leverages the generative adversarial network (GAN) architecture consisting of a generator network and a discriminator network, with a combined loss of the mean squared error (MSE) and binary cross-entropy (BC) losses, for regression tasks. In this work, the generator employs design requirements as input and produces optimal takeoff trajectory profiles, while the discriminator differentiates the generated profiles and real optimal profiles in the training set. The combined loss facilitates the generator training in the dual aspects: the MSE loss targets minimum differences between generated profiles and training counterparts, while the BC loss drives the generated profiles to share analogous patterns with the training set. We demonstrated the utility of regGAN-TL on optimal takeoff trajectory designs for the Airbus A3 Vahana and compared its performance against representative surrogates, including the multi-output Gaussian process, the conditional GAN, and the vanilla regGAN. Results showed that regGAN-TL reached the 99.5% generalization accuracy threshold with only 200 training samples while the best reference surrogate required 400 samples. The 50% reduction in training expense and reduced standard deviations of generalization accuracy achieved by regGAN-TL confirmed its outstanding predictive performance and broad engineering application potential.

Generative Adversarial Networks for Synthetic Data Generation in Finance: Evaluating Statistical Similarities and Quality Assessment

Generating synthetic data is a complex task that necessitates accurately replicating the statistical and mathematical properties of the original data elements. In sectors such as finance, utilizing and disseminating real data for research or model development can pose substantial privacy risks owing to the inclusion of sensitive information. Additionally, authentic data may be scarce, particularly in specialized domains where acquiring ample, varied, and high-quality data is difficult or costly. This scarcity or limited data availability can limit the training and testing of machine-learning models. In this paper, we address this challenge. In particular, our task is to synthesize a dataset with similar properties to an input dataset about the stock market. The input dataset is anonymized and consists of very few columns and rows, contains many inconsistencies, such as missing rows and duplicates, and its values are not normalized, scaled, or balanced. We explore the utilization of generative adversarial networks, a deep-learning technique, to generate synthetic data and evaluate its quality compared to the input stock dataset. Our innovation involves generating artificial datasets that mimic the statistical properties of the input elements without revealing complete information. For example, synthetic datasets can capture the distribution of stock prices, trading volumes, and market trends observed in the original dataset. The generated datasets cover a wider range of scenarios and variations, enabling researchers and practitioners to explore different market conditions and investment strategies. This diversity can enhance the robustness and generalization of machine-learning models. We evaluate our synthetic data in terms of the mean, similarities, and correlations.

Exploring Generative Adversarial Networks for Face Generation

This research has used Generative Adversarial Networks (GANs) for face generation in the area of the research [1]. GANs have demonstrated a high effectiveness in imitation of natural images not only, but also in specific applications [2]. For the succeeding paper, their potential especially on creating human faces are being looked into. The primary goal is to develop a new design and practical training strategy to efficiently produce high-quality facial images [3]. Experimental results confirm the ability of the proposed method to generate artificially intelligent faces and its role in improving the quality of the images.

Prediction of Human Reaching Pose Sequences in Human–Robot Collaboration

Abstract In human–robot collaboration, robots and humans must work together in shared, overlapping, workspaces to accomplish tasks. If human and robot motion can be coordinated, then collisions between robot and human can seamlessly be avoided without requiring either of them to stop work. A key part of this coordination is anticipating humans’ future motion so robot motion can be adapted proactively. In this work, a generative neural network predicts a multi-step sequence of human poses for tabletop reaching motions. The multi-step sequence is mapped to a time-series based on a human speed versus motion distance model. The input to the network is the human’s reaching target relative to current pelvis location combined with current human pose. A dataset was generated of human motions to reach various positions on or above the table in front of the human starting from a wide variety of initial human poses. After training the network, experiments showed that the predicted sequences generated by this method matched the actual recordings of human motion within an L2 joint error of 7.6 cm and L2 link roll–pitch–yaw error of 0.301 rad on average. This method predicts motion for an entire reach motion without suffering from the exponential propagation of prediction error that limits the horizon of prior works.

GACDNet:Mapping winter wheat by generative adversarial cross-domain networks with transformer integration for zero-sample extraction

Accurate extraction of winter wheat area is essential for wheat yield estimation. Remotely sensed images have limited coverage, are taken at different times, from different angles and in different geographical areas, and the spectral information of the same feature varies from image to image. Machine learning methods are very popular in extracting winter wheat planting area, there are differences between different images, namely the distribution differences between source domain data and target domain data, the results obtained by these methods directly applied to other areas are not satisfactory. To achieve cross regional extraction of winter wheat area, we propose generative adversarial cross-domain networks for image classification under the same image type. The cross-domain network proposed comprises a generative network and a feature extractor. The generative network creates diverse and acceptable samples from the initial input data, constrained by the contrast loss. The feature extractor converts the initial input data and the generated data into a high-level representation. Two time series datasets were constructed and a series of experiments were conducted on these datasets, obtaining high-precision and optimal classification results. Sample training was performed using 70%, 50%, 30% and 10% of the data on both datasets. The study results demonstrated the high performance and cross-domain capability of our network. Furthermore, the winter wheat cultivation area of the entire Zhoukou city was extracted, yielding results with an accuracy of 94.23%. Extensive experiments and applications on the dataset demonstrate that our method is feasible and reliable.

Synthetic Data for Deep Learning in Computer Vision & Medical Imaging: A Means to Reduce Data Bias

Deep-learning (DL) performs well in computer-vision and medical-imaging automated decision-making applications. A bottleneck of DL stems from the large amount of labelled data required to train accurate models that generalise well. Data scarcity and imbalance are common problems in imaging applications that can lead DL models towards biased decision making. A solution to this problem is synthetic data. Synthetic data is an inexpensive substitute to real data for improved accuracy and generalisability of DL models. This survey reviews the recent methods published in relation to the creation and use of synthetic data for computer-vision and medical-imaging DL applications. The focus will be on applications that utilised synthetic data to improve DL models by either incorporating an increased diversity of data that is difficult to obtain in real life, or by reducing a bias caused by class imbalance. Computer-graphics software and generative networks are the most popular data generation techniques encountered in the literature. We highlight their suitability for typical computer-vision and medical-imaging applications, and present promising avenues for research to overcome their computational and theoretical limitations.

Few-shot SAR vehicle target augmentation based on generative adversarial networks

Abstract. The study of few-shot SAR image generation is an effective way to expand the SAR dataset, which not only provides diversified data support for SAR target classification, but also provides a high-fidelity false image template for SAR deceptive jamming. In this paper, we have constructed a multi-frequency and multi-target type SAR vehicle imagery dataset that encompasses frequencies such as X, Ka, P, and S bands. The vehicle types are coaster, suv and cabin. Subsequently, we utilized various Generative Adversarial Networks for image generation from the SAR vehicle dataset. The experimental result indicates that the images generated by the DCGAN and the LSGAN models are of superior quality. Furthermore, we employed different recognition networks to evaluate the classification accuracy of the generated images. Of all the frequency bands, the Ka band generated images achieved the highest recognition rate, with an accuracy of up to 99%. Under conditions of a limited number of samples, the LSGAN model performed the best, reaching a classification recognition rate of 71.48% with a dataset of only 20 samples. Finally, we use a conditional network generation model to generate conditions based on target categories and frequency bands, providing high fidelity samples for SAR deception jamming.

A cutting-edge intelligent cyber model for intrusion detection in IoT environments leveraging future generations networks

A cutting-edge intelligent cyber model for intrusion detection in IoT environments leveraging future generations networks

GANSharp: High-definition image reconstruction using generative adversarial networks

In the realm of digital imaging, enhancing low-resolution images to high-definition quality is a pivotal challenge, particularly crucial for applications in medical imaging, security, and remote sensing. Traditional methods, primarily relying on basic interpolation techniques, often result in images that lack detail and fidelity. GANSharp introduces an innovative GAN-based framework that substantially improves the generator network, incorporating adversarial and perceptual loss functions for enhanced image reconstruction. The core issue addressed is the loss of critical information during down-sampling processes. To counteract this, we proposed a GAN-based method leveraging deep learning algorithms, trained using sets of both low- and high-resolution images. Our approach, which focuses on expanding the generator network’s size and depth and integrating adversarial and perceptual loss, was thoroughly evaluated on various benchmark datasets. The experimental results showed remarkable outcomes. On the Set5 dataset, our method achieved a PSNR of 34.18 dB and a SSIM of 0.956. Comparatively, on the Set14 dataset, it yielded a PSNR of 31.16 dB and an SSIM of 0.920, and on the B100 dataset, it achieved a PSNR of 30.51 dB and an SSIM of 0.912. These results were superior or comparable to those of existing advanced algorithms, demonstrating the proposed method’s potential in generating high-quality, high-resolution images. Our research underscores the potency of GANs in image super-resolution, making it a promising tool for applications spanning medical diagnostics, security systems, and remote sensing. Future exploration could extend to the utilization of alternative loss functions and novel training techniques, aiming to further refine the efficacy of GAN-based image restoration algorithms.

A bearing fault diagnosis method based on LSTM-GAN and convolutional neural network under small sample variable working conditions

In order to improve the identification accuracy of bearing fault diagnosis, overcome the training difficulties and poor generalization ability of fault diagnosis model under the condition of small samples, this work constructs the LSTM-GAN model by combining long short-term memory network (LSTM) with generative adductive neural network (GAN). Firstly, LSTM is used to build a generator to generate adversarial neural network model, and the feature extraction capability of LSTM is adopted to improve the quality of generated samples. Then, the convolutional neural network (CNN) is improved to enhance its classification ability, and the improved CNN is used to classify faults. Finally, CNN and convolutional autoencoder (CAE) are used to diagnose bearing faults under different working conditions to enhance the diagnostic effect of the model under different working conditions. The results show that LSTM-GAN can capture the feature information in the original data well, and the generated samples can improve the diagnosis accuracy of bearing fault diagnosis under the condition of small samples. The diagnostic model still has high accuracy under different working conditions, which provides support for the research and application of bearing fault diagnosis.

Edge-guided generative network with attention for point cloud completion

Edge-guided generative network with attention for point cloud completion

Transformer fault diagnosis based on adversarial generative networks and deep stacked autoencoder

Establishing a deep learning model for transformer fault diagnosis using transformer oil chromatogram data requires a large number of fault samples. The lack and imbalance of oil chromatogram data can lead to overfitting, lack of representativeness of the model, and unsatisfactory prediction results on test set data, making it difficult to accurately diagnose transformer faults. A conditional Wasserstein generative adversarial network with gradient penalty optimization (CWGAN-GP) is adopted in this paper, which based on gradient penalty optimization and expand the oil chromatography fault samples of 500 sets of transformer oil chromatography data with 5 types of faults. The proposed method is used to classify transformer faults using a deep autoencoder, and the sample quality of the neural network model proposed in this paper is compared with several other variants of generative adversarial neural network models. The research results show that after using the method proposed in this paper for sample expansion, the overall accuracy of fault diagnosis can reach 93.2 %, which is 4.98 % higher than the original imbalanced samples. Compared with other sample expansion methods, the accuracy of fault diagnosis of the algorithm in this paper is improved by 1.70 %–3.05 %.

3D-DGGAN: A Data-Guided Generative Adversarial Network for High Fidelity in Medical Image Generation.

Three-dimensional images are frequently used in medical imaging research for classification, segmentation, and detection. However, the limited availability of 3D images hinders research progress due to network training difficulties. Generative methods have been proposed to create medical images using AI techniques. Nevertheless, 2D approaches have difficulty dealing with 3D anatomical structures, which can result in discontinuities between slices. To mitigate these discontinuities, several 3D generative networks have been proposed. However, the scarcity of available 3D images makes training these networks with limited samples inadequate for producing high-fidelity 3D images. We propose a data-guided generative adversarial network to provide high fidelity in 3D image generation. The generator creates fake images with noise using reference code obtained by extracting features from real images. The generator also creates decoded images using reference code without noise. These decoded images are compared to the real images to evaluate fidelity in the reference code. This generation process can create high-fidelity 3D images from only a small amount of real training data. Additionally, our method employs three types of discriminator: volume (evaluates all the slices), slab (evaluates a set of consecutive slices), and slice (evaluates randomly selected slices). The proposed discriminator enhances fidelity by differentiating between real and fake images based on detailed characteristics. Results from our method are compared with existing methods by using quantitative analysis such as Fréchet inception distance and maximum mean discrepancy. The results demonstrate that our method produces more realistic 3D images than existing methods.

Identification of propagation path and root cause of faults based on generative adversarial networks in industrial systems

Fault diagnosis and detection (FDD) improves product quality, process safety, and profitability in modern industrial plants. Faults can propagate between different parts of plants due to connections between devices and control loops. As a result, control room operators may receive multiple alarms that distract their attention from addressing the main causes of abnormal situations. Therefore, investigating causal relationships between process variables is essential in determining the root cause of faults. This study proposes a new systematic approach for causality analysis to identify the propagation path and root cause of faults in industrial plants. The research's innovations are categorized into two main parts. Initially, the study utilizes machine learning algorithms to identify causal relationships using probabilistic concepts. Additionally, it introduces a novel architecture by merging the generative adversarial network (GAN) with the variational auto-encoder (VAE). The process variables are mapped to a latent space, where an adversarial generative network will investigate the causal relationship between variables. Finally, the Tennessee Eastman Process simulation and gas turbine industrial data are used to evaluate the proposed algorithm.

Transferable Sparse Adversarial Attack on Modulation Recognition With Generative Networks

Transferable Sparse Adversarial Attack on Modulation Recognition With Generative Networks

Text Feature Adversarial Learning for Text Generation With Knowledge Transfer From GPT2.

Text generation is a key component of many natural language tasks. Motivated by the success of generative adversarial networks (GANs) for image generation, many text-specific GANs have been proposed. However, due to the discrete nature of text, these text GANs often use reinforcement learning (RL) or continuous relaxations to calculate gradients during learning, leading to high-variance or biased estimation. Furthermore, the existing text GANs often suffer from mode collapse (i.e., they have limited generative diversity). To tackle these problems, we propose a new text GAN model named text feature GAN (TFGAN), where adversarial learning is performed in a continuous text feature space. In the adversarial game, GPT2 provides the "true" features, while the generator of TFGAN learns from them. TFGAN is trained by maximum likelihood estimation on text space and adversarial learning on text feature space, effectively combining them into a single objective, while alleviating mode collapse. TFGAN achieves appealing performance in text generation tasks, and it can also be used as a flexible framework for learning text representations.

Chemical fault diagnosis network based on single domain generalization

Recent developments in fault diagnosis have leveraged domain generalization to address the issue of domain shift. Most existing methods focus on learning domain-invariant representations from multiple source domains. However, collecting valuable fault samples from varying operational conditions is challenging, and it is common for available data to originate from a single operational condition. Thus, this paper introduces a Multi-scale generative and adversarial Metric networks (MGAMN) for Chemical Process Fault Diagnosis. To enhance model generalization, a domain generation module was developed to create augmented domains with significant distributional differences from the source domain. The diagnostic task module then extracts domain-invariant features from both the source and augmented domains. A multi-scale generation strategy is established, utilizing multi-scale deep separable convolutions (Dsc) to ensure that the generated samples contain rich state information. Additionally, an adversarial training and metric learning strategy is designed to learn generalized features capable of resisting unknown domain shifts. Extensive diagnostic experiments on the non-isothermal continuous stirred tank reactor (CSTR) and the Tennessee Eastman Process (TEP) chemical datasets validate the effectiveness of the proposed method. Moreover, ablation studies confirm the effectiveness of the proposed modular strategy, demonstrating significant potential for practical applications.