Deep learning (DL) has achieved remarkable successes in many disciplines such as computer vision and natural language processing due to the availability of “big data”. However, such success cannot be ...easily replicated in many nuclear engineering problems because of the limited amount of training data, especially when the data comes from high-cost experiments. To overcome such a data scarcity issue, this paper explores the applications of deep generative models (DGMs) that have been widely used for image data generation to scientific data augmentation. DGMs, such as generative adversarial networks (GANs), normalizing flows (NFs), variational autoencoders (VAEs), and conditional VAEs (CVAEs), can be trained to learn the underlying probabilistic distribution of the training dataset. Once trained, they can be used to generate synthetic data that are similar to the training data and significantly expand the dataset size. By employing DGMs to augment TRACE simulated data of the steady-state void fractions based on the NUPEC Boiling Water Reactor Full-size Fine-mesh Bundle Test (BFBT) benchmark, this study demonstrates that VAEs, CVAEs, and GANs have comparable generative performance with similar errors in the synthetic data, with CVAEs achieving the smallest errors. The findings shows that DGMs have a great potential to augment scientific data in nuclear engineering, which proves effective for expanding the training dataset and enabling other DL models to be trained more accurately.
•This paper explores the applications of deep generative models (DGMs) to scientific data augmentation.•These models offer a potential solution to the data scarcity issue in nuclear engineering.•We tested the performance of several DGMs for data augmentation, namely GANs, real NVP NFs, VAEs and CVAEs.•TRACE simulations of steady-state void fraction dataset in the BFBT benchmark was used for the demonstration.•The synthetic data generated by the four models demonstrated their ability to produce credible samples.
With the high penetration of renewable generation systems in the power grid, the accurate simulation of the uncertainty in renewable energy generation is vital to the safe operation of the power ...system This paper proposes a novel controllable method for renewable scenario generation based on the improved VAEGAN model. The standard VAEGAN model is first improved using spectral normalization technique and the generator of GAN is trained using VAE. Then, the external and internal interpretable features in the latent space are learned as the controllable vector utilizing the principle of mutual information maximization. Finally, the renewable energy scenarios with overall features are generated using the external universal meteorological features, and renewable energy scenarios with specific features are generated by tuning along the internal interpretable feature of the controllable vector in the latent space. The proposed approach is used to produce real-time series data for renewable energy including wind and solar power. Experiments demonstrate that our method has a better performance in terms of controllable generation and enables the generation of preference patterns covering various statistical features.
•A novel VAEGAN model with controllable interpretable features is proposed.•External universal meteorological features are used to generate scenarios with unknown patterns.•Internal interpretable latent features are designed to generate scenarios with specific features.•The results show a better performance of the proposed method on controllable scenario generation.
Many industries are evaluating the use of the Internet of Things (IoT) technology to perform remote monitoring and predictive maintenance on their mission-critical assets and equipment, for which ...mechanical bearings are their indispensable components. Although many data-driven methods have been applied to bearing fault diagnosis, most of them belong to the supervised learning paradigm that requires a large amount of labeled training data to be collected in advance. In practical applications, however, obtaining labeled data that accurately reflect real-time bearing conditions can be more challenging than collecting large amounts of unlabeled data. In this paper, we thus propose a semi-supervised learning scheme for bearing fault diagnosis using variational autoencoder (VAE)-based deep generative models, which can effectively leverage a dataset when only a small subset of data have labels. Finally, a series of experiments were conducted using the University of Cincinnati Intelligent Maintenance System (IMS) Center dataset and the Case Western Reserve University (CWRU) bearing dataset. The experimental results demonstrate that the proposed semi-supervised learning schemes outperformed some mainstream supervised and semi-supervised benchmarks with the same percentage of labeled data samples. Additionally, the proposed methods can mitigate the label inaccuracy issue when identifying naturally-evolved bearing defects.
•The WiFi fingerprint-based positioning problems are investigated.•A hybrid deep learning-based model is introduced.•A deep learning-based semi-supervised model is introduced.•The validation ...experiments are conducted to demonstrate the effectiveness of the proposed methods.
WiFi fingerprint-based indoor localization has been a popular research topic recently. In this work, we propose two novel deep learning-based models, the convolutional mixture density recurrent neural network and the variational autoencoder-based semi-supervised learning model. The convolutional mixture density recurrent neural network is designed for indoor next location prediction, in which the advantages of convolutional neural networks, recurrent neural networks and mixture density networks are combined. Furthermore, since most of real-world WiFi fingerprint data are not labeled, we devise a variational autoencoder-based model to compute accurate user location in a semi-supervised learning manner. Finally, in order to evaluate the proposed models, we conduct the validation experiments on two real-world datasets. The final results are compared to other existing methods and verify the effectiveness of our approaches.
There is a great leap in objective accuracy on image super-resolution, which recently brings a new challenge on image super-resolution with larger up-scaling (e.g. <inline-formula> <tex-math ...notation="LaTeX">4\times </tex-math></inline-formula>) using pixel based distortion for measurement. This causes over-smooth effect which cannot grasp well the perceptual similarity. The advent of generative adversarial networks makes it possible super-resolve a low-resolution image to generate photo-realistic images sharing distribution with the high-resolution images. However, generative networks suffer from problems of mode-collapse and unrealistic sample generation. We propose to perform Image Super-Resolution via Variational AutoEncoders (SR-VAE) learning according to the conditional distribution of the high-resolution images induced by the low-resolution images. Given that the Conditional Variational Autoencoders tend to generate blur images, we add the conditional sampling mechanism to narrow down the latent subspace for reconstruction. To evaluate the model generalization, we use KL loss to measure the divergence between latent vectors and standard Gaussian distribution. Eventually, in order to balance the trade-off between super-resolution distortion and perception, not only that we use pixel based loss, we also use the modified deep feature loss between SR and HR images to estimate the reconstruction. In experiments, we evaluated a large number of datasets to make comparison with other state-of-the-art super-resolution approaches. Results on both objective and subjective measurements show that our proposed SR-VAE can achieve good photo-realistic perceptual quality closer to the natural image manifold while maintain low distortion.
Supervised machine learning algorithms have been widely used in seismic exploration processing, but the lack of labeled examples complicates its application. Therefore, we propose a seismic labeled ...data expansion method based on deep variational Autoencoders (VAE), which are made of neural networks and contains two parts-Encoder and Decoder. Lack of training samples leads to overfitting of the network. We training the VAE with whole seismic data, which is a data-driven process and greatly alleviates the risk of overfitting. The Encoder captures the ability to map the seismic waveform Y to latent deep features z, and the Decoder captures the ability to reconstruct high-dimensional waveform Yˆ from latent deep features z. Later, we put the labeled seismic data into Encoders and get the latent deep features. We can easily use gaussian mixture model to fit the deep feature distribution of each class labeled data. We resample a mass of expansion deep features z∗ according to the Gaussian mixture model, and put the expansion deep features into the decoder to generate expansion seismic data. The experiments in synthetic and real data show that our method alleviates the problem of lacking labeled seismic data for supervised seismic facies analysis.
•A deep learning method for label data extension is proposed.•Taking whole seismic data as the training data alleviates the risk of over-fitting.•The expansion labeled data has a positive effect on seismic waveform classification.
Human infants learn language while interacting with their environment in which their caregivers may describe the objects and actions they perform. Similar to human infants, artificial agents can ...learn language while interacting with their environment. In this work, first, we present a neural model that bidirectionally binds robot actions and their language descriptions in a simple object manipulation scenario. Building on our previous Paired Variational Autoencoders (PVAE) model, we demonstrate the superiority of the variational autoencoder over standard autoencoders by experimenting with cubes of different colours, and by enabling the production of alternative vocabularies. Additional experiments show that the model's channel-separated visual feature extraction module can cope with objects of different shapes. Next, we introduce PVAE-BERT, which equips the model with a pretrained large-scale language model, i.e., Bidirectional Encoder Representations from Transformers (BERT), enabling the model to go beyond comprehending only the predefined descriptions that the network has been trained on; the recognition of action descriptions generalises to unconstrained natural language as the model becomes capable of understanding unlimited variations of the same descriptions. Our experiments suggest that using a pretrained language model as the language encoder allows our approach to scale up for real-world scenarios with instructions from human users.
Display omitted
•Unsupervised machine learning methods correctly identify species-level divergences.•Multiple empirical and simulated datasets demonstrate their utility.•These approaches are ...versatile, with great potential in integrative species delimitation.•Taxa with high population structure should be considered when testing new methods.
One major challenge to delimiting species with genetic data is successfully differentiating population structure from species-level divergence, an issue exacerbated in taxa inhabiting naturally fragmented habitats. Many fields of science are now using machine learning, and in evolutionary biology supervised machine learning has recently been used to infer species boundaries. These supervised methods require training data with associated labels. Conversely, unsupervised machine learning (UML) uses inherent data structure and does not require user-specified training labels, potentially providing more objectivity in species delimitation. In the context of integrative taxonomy, we demonstrate the utility of three UML approaches (random forests, variational autoencoders, t-distributed stochastic neighbor embedding) for species delimitation in an arachnid taxon with high population genetic structure (Opiliones, Laniatores, Metanonychus). We find that UML approaches successfully cluster samples according to species-level divergences and not high levels of population structure, while model-based validation methods severely over-split putative species. UML offers intuitive data visualization in two-dimensional space, the ability to accommodate various data types, and has potential in many areas of systematic and evolutionary biology. We argue that machine learning methods are ideally suited for species delimitation and may perform well in many natural systems and across taxa with diverse biological characteristics.
Given a multitude of genetic and environmental factors, when investigating the variability in schizophrenia (SCZ) and the first-degree relatives (R-SCZ), latent disease-specific variation is usually ...hidden. To reliably investigate the mechanism underlying the brain deficits from the aspect of functional networks, we newly iterated a framework of contrastive variational autoencoders (cVAEs) applied in the contrasts among three groups, to disentangle the latent resting-state network patterns specified for the SCZ and R-SCZ. We demonstrated that the comparison in reconstructed resting-state networks among SCZ, R-SCZ, and healthy controls (HC) revealed network distortions of the inner-frontal hypoconnectivity and frontal-occipital hyperconnectivity, while the original ones illustrated no differences. And only the classification by adopting the reconstructed network metrics achieved satisfying performances, as the highest accuracy of 96.80% ± 2.87%, along with the precision of 95.05% ± 4.28%, recall of 98.18% ± 3.83%, and F1-score of 96.51% ± 2.83%, was obtained. These findings consistently verified the validity of the newly proposed framework for the contrasts among the three groups and provided related resting-state network evidence for illustrating the pathological mechanism underlying the brain deficits in SCZ, as well as facilitating the diagnosis of SCZ.
•A new framework of cVAEs is proposed and applied the contrasts among three groups.•Inner-frontal hypoconnectivity and frontal-occipital hyperconnectivity is found in SCZ.•Reconstructed networks acquired satisfying classification performance among the three groups.
Videos represent the primary source of information for surveillance applications. Video material is often available in large quantities but in most cases it contains little or no annotation for ...supervised learning. This article reviews the state-of-the-art deep learning based methods for video anomaly detection and categorizes them based on the type of model and criteria of detection. We also perform simple studies to understand the different approaches and provide the criteria of evaluation for spatio-temporal anomaly detection.