Data Augmentation Classifier for Imbalanced Fault Classification

Fuzzy decision fusion system for fault classification with analytic hierarchy process approach

Switching LDS-based approach for process fault detection and classification

This paper considers the problem of fault classification when different fault classifiers are performed simultaneously. Based on spectral meta-learner (SML) proposed by Parisi et al., its recursive version, i.e., recursive SML (RSML) is developed for online combining the potentially conflicting classification information. Considering different statistical properties of faults occurring at different time intervals, the binary classification information is recursively utilized. By introducing a forgetting factor, the leading eigenvector of the estimate of the time-varying covariance matrix is used as the weight vector for each classifier. Rank-one modification is then used for calculating the eigenvector in order to reduce the online computational complexity. The performance of RSML is strictly analyzed in a statistical sense, including the effect of recursive calculation and conditional dependence of different classifiers on the weight vector. Compared with majority voting and SML, a higher balanced accuracy of RSML can be verified by the benchmark Tennessee Eastman process.

Deep learning methods have made remarkable strides in surface defect detection. But, they heavily rely on large amount of training data, which can be a costly endeavor, especially for specific applications like steel strip surface defect detection, where acquiring and labeling large-scale data is impractical due to the rarity of certain defective categories in production environment. Hence, realistic defect image synthesis can greatly alleviate this issue. However, training image generation networks also demand substantial data, making image data augmentation merely an auxiliary effort. In this work, we propose a Generative Adversarial Network (GAN)-based image synthesis framework. We selectively extract the defect edges of the original image as well as the background texture information, and use them as network input through the spatially-adaptive (de)normalization (SPADE) module. This enriches the input information, thus significantly reducing the amount of training data for GAN network in image generation, and enhancing the background details as well as the defect boundaries in the generated images. Additionally, we introduce a novel generator loss term that balances the similarity and perceptual fidelity between synthetic and real images by constraining high-level features at different feature levels. This provides more valuable information for data augmentation in training object detection models using synthetic images. Our experimental results demonstrate the sophistication of the proposed image synthesis method and its effectiveness in data augmentation for steel strip surface defect detection tasks.

This dataset is an extended version of from the aboved paper. The dataset contains more than 500,000 legal arguments which were automatically labeled by combining the following algorithms with majority vote:\n\n\n\tFine-Tuned Legal BERT on training data involving few Paraphrased test samples. Accuracy: 0.834933 \n\tFine-Tuned Legal BERT with custom algorithm involving a corruption of few train samples. Accuracy: 0.76775 \n\tFine-Tuned Legal BERT Accuracy: 0.76199 \n\tFine-Tuned BERT Accuracy: 0.76007 \n\tGAN-BERT Accuracy: 0.7480\n\tCo Training Accuracy: 0.66410 \n\tEM-LightGBM Accuracy: 0.6564 \n\tPseudo Labeling Accuracy: 0.6339 \n\n\nThere are six labels in the dataset: 1) fact, 2) issue, 3)rule/law/holding, 4) analysis, 5) conclusion/opinion/answer, and 6) others. For more details about the labels and the algorithms, please read our paper on: IEEE Transactions on Reliability

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.

Recently, in the building and infrastructure fields, studies on defect detection methods using deep learning have been widely implemented. For robust automatic recognition of defects in buildings, a sufficiently large training dataset is required for the target defects. However, it is challenging to collect sufficient data from degrading building structures. To address the data shortage and imbalance problem, in this study, a data augmentation method was developed using a generative adversarial network (GAN). To confirm the effect of data augmentation in the defect dataset of old structures, two scenarios were compared and experiments were conducted. As a result, in the models that applied the GAN-based data augmentation experimentally, the average performance increased by approximately 0.16 compared to the model trained using a small dataset. Based on the results of the experiments, the GAN-based data augmentation strategy is expected to be a reliable alternative to complement defect datasets with an unbalanced number of objects.

Deep Learning algorithms have almost become a key standard for majority of vision and machine learning problems. Despite its common usage and high performance for many applications, they have certain disadvantages. One major problem with deep learning methods is the size of the dataset to be used for training. The methods require a large dataset for an adequate training. However, a large dataset may not be available for all problems. In such a case, researchers apply data augmentation methods to obtain a larger dataset from a given dataset. For the image classification problem, conventional method for data augmentation is the application of transformation based methods; such as flipping, rotation, blurring etc. Recently, generative models which apply deep learning methods are also commonly used for data augmentation. On the other hand, in case of a too large dataset the classifiers may overfit and end up with a lack of generalization. In this study, we explore the usage of generative adversarial networks for data augmentation in the image classification problem. We evaluate the classification performance with three types of augmentation methods. Dataset is first augmented by two conventional methods; Gaussian blurring and dropout of regions, then by generative adversarial networks. Meanwhile, we observe the behavior of the classifier for various sized datasets with and without data augmentation. We observe that especially in datasets of certain sizes generative adversarial networks can be effectively used for data augmentation.

Traditional data-driven intelligent fault diagnosis methods for rotating component commonly assume that sufficient labeled data is available. However, the rotary machine works in a normal state most of the time in practical engineering, resulting in fault diagnosis scenarios with extremely limited and imbalanced data. This phenomenon deteriorates the performance of intelligent fault diagnosis methods. In this article, a data augmentation method based on an improved variational autoencoding generative adversarial network is proposed to address this problem. First, the self-attention mechanism and Wasserstein distance with gradient penalty are introduced to improve data generation capability and training stability of the generative model. Then, the generative models are used to generate synthetic samples with high similarity and diversity based on extremely limited frequency spectrum samples. Subsequently, data filters are established to automatically refine the generated samples. Finally, the refined samples are used to enrich imbalanced dataset and improve the fault diagnosis accuracy. The effectiveness of the proposed method is validated by vibration datasets of two rotating components. Experimental results show that the proposed method can generate high-quality samples and has the potential to enhance the fault diagnosis performance in scenarios with limited data.

Machine learning (ML) has been extensively adopted for online sensing-based monitoring in advanced manufacturing systems. However, the sensor data collected under abnormal states are usually insufficient, leading to significant data imbalance issues for supervised ML. A common solution is to incorporate data augmentation techniques; that is, augmenting the available abnormal states data (i.e., minority samples) via synthetic generation. To generate the high-quality minority samples, it is vital to learn the underlying distribution of the abnormal states data. In recent years, generative adversarial network (GAN)–based approaches have become popular to learn data distribution, as well as perform data augmentation. However, in practice, the quality of generated samples from GAN-based data augmentation may vary drastically. In addition, the sensor signals are collected sequentially by time from the manufacturing systems, which means sequential information is also very important in data augmentation. To address these limitations, inspired by the multihead attention mechanism, this article proposed an attention-stacked GAN (AS-GAN) architecture for sensor data augmentation of online monitoring in manufacturing systems. It incorporates a new attention-stacked framework to strengthen the generator in GAN with the capability of capturing sequential information, and thereby the developed attention-stacked framework greatly helps to improve the quality of the generated sensor signals. Afterward, the generated high-quality sensor signals for abnormal states could be applied to train classifiers more accurately, further improving the online monitoring performance of manufacturing systems. The case study conducted in additive manufacturing also successfully validated the effectiveness of the proposed AS-GAN.

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.

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.

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.

<p>High data collection costs and complicated data access regulations increase the demand for synthetic data. Generative Adversarial Networks (GANs) are a novel generative framework with great potential for high quality synthetic data generation. GANs formulate the true distribution of data implicitly, and the success of GANs are often measured based on the similarity of generated data to this true distribution. GANs were originally designed to work with continuous data. However, many important real-world datasets such as medical images involve discontinuous distributions. GAN training for discontinuous distributions is relatively more challenging, as the training procedure often suffers from instability and mode collapse issues. This dissertation focuses on designing novel GAN architectures to generate representative synthetic image data, and proposes new structures to alleviate GANs' mode collapse issue. As part of this thesis, novel applications of image data generation with GANs have been also investigated for important problems arising in the telecommunication industry and medical domain. Specifically, we first explore various GAN structures to generate engineered electromagnetic surfaces. We consider the continuous approximation of the data and explore the capabilities of feed-forward and convolutional GANs for synthetic data generation. Next, we introduce a novel GAN architecture to address the problem of mode collapse in GAN training. The proposed structure incorporates a third network that penalizes the generator for generating low diversity samples. Lastly, we study the challenging problem of object generation in 3D space using GANs, and we propose extensions to existing 3D GAN structures to generate connected 3D volumes. Additionally, we explore a more challenging version of this 3D volume generation problem by generating connected volumes packed with spheres. This research has applications in radiosurgery treatment planning, and the proposed 3D GAN structure can help generate rare, unseen 3D tumor volumes and information on how to treat these tumors. Accordingly, our analysis contributes to overcoming data scarcity issues (e.g., due to privacy considerations) for an important practical problem in the medical domain.</p>

Due to the real working conditions, the collected mechanical fault datasets are actually limited and always highly imbalanced, which restricts the diagnosis accuracy and stability. To solve these problems, we present an imbalanced fault diagnosis method based on the generative model of conditional-deep convolutional generative adversarial network (C-DCGAN) and provide a study in detail. Deep convolutional generative adversarial network (DCGAN), based on traditional generative adversarial networks (GAN), introduces convolutional neural network into the training for unsupervised learning to improve the effect of generative networks. Conditional generative adversarial network (CGAN) is a conditional model obtained through introducing conditional extension into GAN. C-DCGAN is a combination of DCGAN and CGAN. In C-DCGAN, based on the feature extraction ability of convolutional networks, through the structural optimization, conditional auxiliary generative samples are used as augmented data and applied in machine fault diagnosis. Two datasets (Bearing dataset and Planetary gear box dataset) are carried out to validate. The simulation experiments showed that the improved performance is mainly due to the generated signals from C-DCGAN to balance the dataset. The proposed method can deal with imbalanced fault classification problem much more effectively. This model could improve the accuracy of fault diagnosis and the generalization ability of the classifier in the case of small samples and display better fault diagnosis performance.