Sound Model Factory: An Integrated System Architecture for Generative Audio Modelling

In this paper, we propose an improved hybrid approach for text generation that combines the strengths of generative adversarial networks (GANs) and recurrent neural networks (RNNs). The proposed model, named TextGen-GAN, uses a GAN to generate high-level semantic representations of text, which are then passed through an RNN decoder to produce coherent and diverse text. The suggested method produces superior results in terms of both quality and diversity when tested on a number of benchmark datasets. The obtained results show that the combination of GAN and RNN outperforms GAN alone, RNN only, baseline, and state-of-the-art models, demonstrating the effectiveness of the proposed hybrid approach.

The article presents an analysis of neural networks that can be used to generating web page program code from input image. As part of the research of the possibilities of modification of algorithms for generating code, the method of generating code simultaneous using the generative adversarial and recurrent neural networks based on the input image is proposed. Methods for creating source code using generative adversarial network (GAN) and recurrent neural networks (RNN) are analyzed. Using domain-specific language (DSL) for tree generation code method, it was possible to use GAN and RNN together, achieving on the average 10% productivity increase. Using GRID’s mesh, the generated code is adaptable. To implement the proposed method, software with a graphical user interface is created. Classes of GAN and RNN were considered, among which the feasibility of using a neural network with long-term short-term memory (LSTM) for analysis and forecasting of tags, blocks and HTML elements of the language in the input image was selected and substantiated. The stages of the research are presented. For the experiment datasets made on the basis of existing sites from open sources for 2018–2019 are used. The estimation of the effectiveness of the proposed method is described.

Sonography synthesis has a wide range of applications, including medical procedure simulation, clinical training and multimodality image registration. In this paper, we propose a machine learning approach to simulate ultrasound images at given 3D spatial locations (relative to the patient anatomy), based on conditional generative adversarial networks (GANs). In particular, we introduce a novel neural network architecture that can sample anatomically accurate images conditionally on spatial position of the (real or mock) freehand ultrasound probe. To ensure an effective and efficient spatial information assimilation, the proposed spatially-conditioned GANs take calibrated pixel coordinates in global physical space as conditioning input, and utilise residual network units and shortcuts of conditioning data in the GANs' discriminator and generator, respectively. Using optically tracked B-mode ultrasound images, acquired by an experienced sonographer on a fetus phantom, we demonstrate the feasibility of the proposed method by two sets of quantitative results: distances were calculated between corresponding anatomical landmarks identified in the held-out ultrasound images and the simulated data at the same locations unseen to the networks; a usability study was carried out to distinguish the simulated data from the real images. In summary, we present what we believe are state-of-the-art visually realistic ultrasound images, simulated by the proposed GAN architecture that is stable to train and capable of generating plausibly diverse image samples.

In recent years, generative adversarial network (GAN) has attracted great attention due to its impressive performance and potential numerous applications, such as data augmentation, real-like image synthesis, image compression improvement, etc. The generator in GAN learns the density of the distribution from real data in order to generate high fidelity fake samples from latent space and deceive the discriminator. Despite its advantages, GAN can easily memorize training samples because of the high model complexity of deep neural networks. Thus, training a GAN with sensitive or private data samples may compromise the privacy of training data. To address this privacy issue, we propose a novel \textit{Privacy Preserving Generative Adversarial Network} (PPGAN) that perturbs the objective function of discriminator by injecting Laplace noises based on functional mechanism to guarantee the differential privacy of training data. Since generator training is considered as a post-processing step while guaranteeing differential privacy of discriminator, the trained generator should be differentially private to effectively protect data samples. Through detailed privacy analysis, we theoretically prove that PPGAN can provide such strict differential privacy guarantee. With extensive simulation study on the benchmark dataset MNIST, we show the efficacy of the proposed PPGAN under practical privacy budgets.

Magnetic Resonance Imaging (MRI) is one of the most commonly used modalities in medical imaging, and one of the main limitations is its slow scanning and reconstruction speed. Recently, deep learning used for compressed sensing (CS) methods have been proposed to accelerate the acquisition by undersampling in the K-space and reconstruct images with neural networks. However, there are still some challenges remained: First, directly training networks based on L1/L2 distance to the target fully sampled images may lead to fuzzy reconstruction images because L1/L2 loss only enforces the overall image or patch similarity, but does not consider the local details such as anatomical sharpness. Second, Generative Adversarial Networks (GAN) can partially solve this problem. The undersampling image gets the latent space through the encoder, and the image is reconstructed by the decoder based on GAN loss, but it may generate unrealistic details by lacking of constraints in K-space domain. Third, most of the networks after training are fixed and have limited adaptation capability in the inference time, and the patient-specific information cannot be effectively used. To resolve these challenges, we proposed a new compressed sensing GAN reconstruction method, and there are two main contributions: (1) We proposed a encoder-decoder structure, which guided GAN optimization strategy data-consistency in latent space to improve the reconstruction quality such as preserving more local details and improving the anatomical sharpness while constraining GAN to follow the data distribution in the K-space to prevent the unrealistic details. (2) An online update strategy is used to find the best representation in the latent space for the underlying patient, and the reconstruct result can be further improved by incorporating the patient-specific information. Extensive experimental results show the effectiveness of our method. KeywordsMRI reconstructionData-consistencyLatent spaceOnline update

The lack of the sufficient and diverse training data is one of the main challenges limiting performances of the machine learning enabled applications in optical networks. Here, we propose a deep learning based sequential data augmentation technique for the aggregate traffic data augmentation for diverse optical network scenarios. A generative adversarial network (GAN) model is trained with the experimental traffic data to automatically extract the substantial characteristics of the experimental traffic data through the zero-sum game theory and then augment the traffic data adaptively. The statistical evaluation parameters of the augmented traffic are mean, variance and Hurst exponent. To add comparisons, two other classical generative models including the statistical parameter configuration (SPC) model and the variational autoencoder (VAE) model are also adopted to generate the traffic data that are similar to the actual traffic data. The comprehensive comparisons among the proposed GAN, the SPC and VAE show that the performances of the GAN exceed those of the SPC and the VAE obviously. The mean and the variance of the augmented traffic data from the GAN are almost equal to those of the experimental traffic data, where the average deviations are both within 2%. The Hurst exponent of the augmented traffic data from the GAN is respectively near 90% and 96% of those of the experimental traffic data in the access network and the core network. To estimate the similarity intuitively, the well-known k-mean algorithm is used to cluster the augmented traffic data according to the centroids determined by the corresponding experimental traffic data and the clustering accuracies are all higher than 95% for 6 kinds of typical traffic types in the optical networks. These results demonstrate that the proposed GAN is able to effectively generate the traffic data that is very close to the experimental traffic data and is difficult to be distinguished for diverse traffic types. Moreover, a relatively small dataset with a few hundred pieces of experimental traffic data is required and the amount of the augmented traffic data from the GAN is unlimited in theory, which can be augmented as much as we need. The proposed traffic data augmentation technique also has the potential to be utilized in other sequential data augmentation applications for the optical networks.

In recent years, Deep Learning (DL), such as the algorithms of Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN) and Generative Adversarial Networks (GAN), has been widely studied and applied in various fields including agriculture. Researchers in the fields of agriculture often use software frameworks without sufficiently examining the ideas and mechanisms of a technique. This article provides a concise summary of major DL algorithms, including concepts, limitations, implementation, training processes, and example codes, to help researchers in agriculture to gain a holistic picture of major DL techniques quickly. Research on DL applications in agriculture is summarized and analyzed, and future opportunities are discussed in this paper, which is expected to help researchers in agriculture to better understand DL algorithms and learn major DL techniques quickly, and further to facilitate data analysis, enhance related research in agriculture, and thus promote DL applications effectively. Keywords: deep learning, smart agriculture, neural network, convolutional neural networks, recurrent neural networks, generative adversarial networks, artificial intelligence, image processing, pattern recognition DOI: 10.25165/j.ijabe.20181104.4475 Citation: Zhu N Y, Liu X, Liu Z Q, Hu K, Wang Y K, Tan J L, et al. Deep learning for smart agriculture: Concepts, tools, applications, and opportunities. Int J Agric & Biol Eng, 2018; 11(4): 32-44.

Due to the high acoustic variability, children speech recognition suffers significant performance reduction on most ASR systems which are optimized mainly using adults speech with limited or even none children speech. One of the most straight ideas to solve this problem is to increase the children's speech data during training, however, it is restricted by the more difficult process and higher cost when collecting children's speech compared to adults'. In this work, we develop a generative adversarial network (GANs) based data augmentation method to increase the size of children's training data to improve speech recognition performance for children's speech. Two different types of GANs are explored under WGAN-GP training framework, including the unconditional GANs with an unsupervised learning framework and the conditional GANs using acoustic states as conditions. The proposed data augmentation approaches are evaluated on a Mandarin speech recognition task, with only 40-hour children speech or further including 100-hour adult speech in the training. The results show that more than relative 20% WER reduction can be obtained on children speech testset with the proposed method, and the generated children speech with GAN even can improve the adults' speech within our experimental setups.

In the field of deep learning, the generative model did not attract much attention until GANs (generative adversarial networks) appeared. In 2014, Google’s Ian Goodfellow proposed a generative model called GANs. GANs use different structures and objective functions from the existing generative model. For example, GANs use two neural networks: a generator that creates a realistic image, and a discriminator that distinguishes whether the input is real or synthetic. If there are no problems in the training process, GANs can generate images that are difficult even for experts to distinguish in terms of authenticity. Currently, GANs are the most researched subject in the field of computer vision, which deals with the technology of image style translation, synthesis, and generation, and various models have been unveiled. The issues raised are also improving one by one. In image synthesis, BEGAN (Boundary Equilibrium Generative Adversarial Network), which outperforms the previously announced GANs, learns the latent space of the image, while balancing the generator and discriminator. Nonetheless, BEGAN also has a mode collapse wherein the generator generates only a few images or a single one. Although BEGAN-CS (Boundary Equilibrium Generative Adversarial Network with Constrained Space), which was improved in terms of loss function, was introduced, it did not solve the mode collapse. The discriminator structure of BEGAN-CS is AE (AutoEncoder), which cannot create a particularly useful or structured latent space. Compression performance is not good either. In this paper, this characteristic of AE is considered to be related to the occurrence of mode collapse. Thus, we used VAE (Variational AutoEncoder), which added statistical techniques to AE. As a result of the experiment, the proposed model did not cause mode collapse but converged to a better state than BEGAN-CS.

Generative adversarial networks (GANs) create images by pitting a generator (G) against a discriminator (D) network, aiming to find a balance between the networks. However, achieving this balance is difficult because G is trained based on just one value representing D’s prediction, and only D can access image features. We introduce a novel approach for training GANs using explainable artificial intelligence (XAI) to enhance the quality and diversity of generated images in histopathological datasets. We leverage XAI to extract feature information from D and incorporate it into G via the loss function, a unique strategy not previously explored in this context. We demonstrate that this approach enriches the training with relevant information and promotes improved quality and more variability in the artificial images, decreasing the FID by up to 32.7% compared to traditional methods. In the data augmentation task, these images improve the classification accuracy of Transformer models by up to 3.81% compared to models without data augmentation and up to 3.01% compared to traditional GAN data augmentation. The Saliency method provides G with the most informative feature information. Overall, our work highlights the potential of XAI for enhancing GAN training and suggests avenues for further exploration in this field.

Generative adversarial networks (GANs) have shown their superiority for speech enhancement. Nevertheless, most previous attempts had convolutional layers as the backbone, which may obscure long-range dependencies across an input sequence due to the convolution operator’s local receptive field. One popular solution is substituting recurrent neural networks (RNNs) for convolutional neural networks, but RNNs are computationally inefficient, caused by the unparallelization of their temporal iterations. To circumvent this limitation, we propose an end-to-end system for speech enhancement by applying the self-attention mechanism to GANs. We aim to achieve a system that is flexible in modeling both long-range and local interactions and can be computationally efficient at the same time. Our work is implemented in three phases: firstly, we apply the stand-alone self-attention layer in speech enhancement GANs. Secondly, we employ locality modeling on the stand-alone self-attention layer. Lastly, we investigate the functionality of the self-attention augmented convolutional speech enhancement GANs. Systematic experiment results indicate that equipped with the stand-alone self-attention layer, the system outperforms baseline systems across classic evaluation criteria with up to 95% fewer parameters. Moreover, locality modeling can be a parameter-free approach for further performance improvement, and self-attention augmentation also overtakes all baseline systems with acceptably increased parameters.

Generative models rely on the idea that data can be represented in terms of latent variables which are uncorrelated by definition. Lack of correlation among the latent variable support is important because it suggests that the latent-space manifold is simpler to understand and manipulate than the real-space representation. Many types of generative model are used in deep learning, e.g., variational autoencoders (VAEs) and generative adversarial networks (GANs). Based on the idea that the latent space behaves like a vector space Radford et al. (2015), we ask whether we can expand the latent space representation of our data elements in terms of an orthonormal basis set. Here we propose a method to build a set of linearly independent vectors in the latent space of a trained GAN, which we call quasi-eigenvectors. These quasi-eigenvectors have two key properties: i) They span the latent space, ii) A set of these quasi-eigenvectors map to each of the labeled features one-to-one. We show that in the case of the MNIST image data set, while the number of dimensions in latent space is large by design, 98% of the data in real space map to a sub-domain of latent space of dimensionality equal to the number of labels. We then show how the quasi-eigenvectors can be used for Latent Spectral Decomposition (LSD). We apply LSD to denoise MNIST images. Finally, using the quasi-eigenvectors, we construct rotation matrices in latent space which map to feature transformations in real space. Overall, from quasi-eigenvectors we gain insight regarding the latent space topology.

Hydrogenerators are strategic assets for power utilities. Their reliability and availability can lead to significant benefits. For decades, monitoring and diagnosis of hydrogenerators have been at the core of maintenance strategies. The main cause of hydrogenerator breakdown comes from failure of its high voltage stator, which is a major component of hydrogenerators. In our previous study, it was shown that more than 85% of stator failure mechanisms indicate the presence of Partial Discharges (PD) activity. PD are minute sparks that occur within voids inside high voltage insulation or in the air around the insulating system. Each PD event does not cause immediate failure, but it will slowly erode the insulation system and will lead to breakdown in years to decades. PD signal can be detected from the lead of the hydrogenerator while it is running, thus allowing for on-line diagnosis. Hydro-Québec has been collecting more than 33.000 unlabeled PD measurement files over the last decades including two types of measurement instruments. One of the instruments used is the Partial Discharge Analyzer (PDA). PDs have several different sources and each source is characterized by a specific signature. It is therefore essential to be able to automatically recognize the nature of the DPs in order to anticipate possible degradation. Expert rules to characterize these PD signals exist but these rules cannot be used as an automatic classification tool. Indeed, for certain ambiguous cases (conflict between classes) or the presence of several signatures at the same time (multi-classes), the expert's judgement remains essential.
 A Variationel AutoEncoder (VAE) is used for dimension reduction and projection into a 2D latent space to analyze the training data and then classify it. The 2D latent space of the VAE allows the data space to be restructured and reorganized to anticipate the performance of the classifier. The problem is how to optimize this latent space and thus obtain the best distribution of the different sources of PDs to maximize the chances of good responses from the classifier? To our knowledge, there is currently no method in the literature that clearly answers this question. What is certain, however, is that the optimization of the VAE learning is directly related to the quality of the learning base, i.e. a large size and a perfectly balanced database (all the cases are present and sufficiently represented). The objective of this paper is to compare the quality of the latent space obtained from the experts' rules with a latent space obtained directly from the input signal in an « End-to-End » approach.
 The first method concerns an original unsupervised deep learning method for PD recognition. Instead of labelling the PD measurement files by the expert for a supervised learning process, we use the rules developed by the experts of Hydro-Québec to create a feature vector from recognizable PD signatures. Indeed, labelling a sufficient quantity of signatures for a supervised approach is very time-consuming and therefore cannot be implemented. This is a common problem in the industry where more and more operational data is available but cannot be labelled by experts, who are busy with other tasks. Expert knowledge is then injected into a characteristic and feature vector. In the second method, several Generative Adversarial Networks (GANs) associated with several types of PDs are thus used to generate artificial signals in order to increase the size of the learning base and especially to balance it. A new latent space is thus obtained from the learning of the VAE exclusively carried out on data generated by the GANs.
 Validation tests based on the 33,000 measurements with metrics for evaluating the performances of the various latent spaces are used.

Generative adversarial networks(GANs) have been successfully applied for generating high quality natural images and have been extended to the generation of RGB videos and 3D volume data. In this paper we consider the task of generating RGB-D videos, which is less extensively studied and still challenging. We explore deep GAN architectures suitable for the task, and develop 4 GAN architectures based on existing video-based GANs. With a facial expression database, we experimentally find that an extended version of the motion and content decomposed GANs, known as MoCoGAN, provides the highest quality RGB-D videos. We discuss several applications of our GAN to content creation and data augmentation, and also discuss its potential applications in behavioral experiments.

Surface defect recognition (SDC) is essential in intelligent manufacturing. Deep learning (DL) is a research hotspot in SDC. Limited defective samples are available in most real-world cases, which poses challenges for DL methods. Given such circumstances, generating defective samples by generative adversarial networks (GANs) is applied. However, insufficient samples and high-frequency texture details in defects make GANs very hard to train, yield mode collapse, and poor image quality, which can further impact SDC. To solve these problems, this article proposes a new GAN called contrastive GAN, which can be trained to generate diverse defects with only extremely limited samples. Specifically, a shared data augmentation (SDA) module is proposed for avoiding overfitting. Then, a feature attention matching (FAM) module is proposed to align features for improving the quality of generated images. Finally, a contrastive loss based on hypersphere is employed to constrain GANs to generate images that differ from the traditional transform. Experiments show that the proposed GAN generates defective images with higher quality and lower variance between real defects compared to other GANs. Synthetic images contribute to pretrained DL networks with accuracies of up to 95.00%–99.56% for Northeastern University (NEU) datasets of different sizes and 91.84% for printed circuit board (PCB) cases, which proves the effectiveness of the proposed method.