Bearing Fault Diagnosis Based on Deep Discriminative Adversarial Domain Adaptation Neural Networks

Intelligent diagnosis driven by big data for mechanical fault is an important means to ensure the safe operation ofequipment. In these methods, deep learning-based machinery fault diagnosis approaches have received increasingattention and achieved some results. It might lead to insufficient performance for using transfer learning alone andcause misclassification of target samples for domain bias when building deep models to learn domain-invariantfeatures. To address the above problems, a deep discriminative adversarial domain adaptation neural networkfor the bearing fault diagnosis model is proposed (DDADAN). In this method, the raw vibration data are firstlyconverted into frequency domain data by Fast Fourier Transform, and an improved deep convolutional neuralnetwork with wide first-layer kernels is used as a feature extractor to extract deep fault features. Then, domaininvariant features are learned from the fault data with correlation alignment-based domain adversarial training.Furthermore, to enhance the discriminative property of features, discriminative feature learning is embeddedinto this network to make the features compact, as well as separable between classes within the class. Finally, theperformance and anti-noise capability of the proposedmethod are evaluated using two sets of bearing fault datasets.The results demonstrate that the proposed method is capable of handling domain offset caused by differentworkingconditions and maintaining more than 97.53% accuracy on various transfer tasks. Furthermore, the proposedmethod can achieve high diagnostic accuracy under varying noise levels.

Feature Selection Method Based on Class Discriminative Degree for Intelligent Medical Diagnosis

By using efficient and timely medical diagnostic decision making,clinicians can positively impact the quality and cost of medical care.However,the high similarity of clinical manifestations between diseases and the limitation of clinicians’knowledge both bring much difficulty to decision making in diagnosis.Therefore,building a decision support system that can assist medical staff in diagnosing and treating diseases has lately received growing attentions in the medical domain.In this paper,we employ a multi-label classification framework to classify the Chinese electronic medical records to establish corresponding relation between the medical records and disease categories,and compare this method with the traditional medical expert system to verify the performance.To select the best subset of patient features,we propose a feature selection method based on the composition and distribution of symptoms in electronic medical records and compare it with the traditional feature selection methods such as chi-square test.We evaluate the feature selection methods and diagnostic models from two aspects,false negative rate(FNR)and accuracy.Extensive experiments have conducted on a real-world Chinese electronic medical record database.The evaluation results demonstrate that our proposed feature selection method can improve the accuracy and reduce the FNR compare to the traditional feature selection methods,and the multi-label classification framework have better accuracy and lower FNR than the traditional expert system.

Predicting Lung Cancers Using Epidemiological Data:A Generative-Discriminative Framework

Predictive models for assessing the risk of developing lung cancers can help identify high-risk individuals with the aim of recommending further screening and early intervention.To facilitate pre-hospital self-assessments,some studies have exploited predictive models trained on non-clinical data(e.g.,smoking status and family history).The performance of these models is limited due to not considering clinical data(e.g.,blood test and medical imaging results).Deep learning has shown the potential in processing complex data that combine both clinical and non-clinical information.However,predicting lung cancers remains difficult due to the severe lack of positive samples among follow-ups.To tackle this problem,this paper presents a generative-discriminative framework for improving the ability of deep learning models to generalize.According to the proposed framework,two nonlinear generative models,one based on the generative adversarial network and another on the variational autoencoder,are used to synthesize auxiliary positive samples for the training set.Then,several discriminative models,including a deep neural network(DNN),are used to assess the lung cancer risk based on a comprehensive list of risk factors.The framework was evaluated on over 55000 subjects questioned between January 2014 and December 2017,with 699 subjects being clinically diagnosed with lung cancer between January 2014 and August 2019.According to the results,the best performing predictive model built using the proposed framework was based on DNN.It achieved an average sensitivity of 76.54%and an area under the curve of 69.24%in distinguishing between the cases of lung cancer and normal cases on test sets.

Entity Burst Discriminative Model for Cumulative Citation Recommendation

Knowledge base acceleration-cumulative citation recommendation(KBA-CCR)aims to detect citation-worthiness documents from a chronological stream corpus for a set of target entities in a knowledge base.Most previous works only consider a number of semantic features between documents and target entities in the knowledge base,and then use powerful machine learning approaches such as logistic regression to classify relevant documents and non-relevant documents.However,the burst activities of an entity have been proved to be a significant signal to predict potential citations.In this paper,an entity burst discriminative model(EBDM)is presented to substantially exploit such burst features.The EBDM presents a new temporal representation based on the burst features,which can capture both temporal and semantic correlations between entities and documents.Meanwhile,in contrast to the bag-of-words model,the EBDM can significantly decrease the number of non-zero entries of feature vectors.An extensive set of experiments were conducted on the TREC-KBA-2012 dataset.The results show that the EBDM outperforms the performance of the state-of-the-art models.

Learning a Discriminative Feature Attention Network for pancreas CT segmentation

Accurate pancreas segmentation is critical for the diagnosis and management of diseases of the pancreas. It is challenging to precisely delineate pancreas due to the highly variations in volume, shape and location. In recent years, coarse-to-fine methods have been widely used to alleviate class imbalance issue and improve pancreas segmentation accuracy. However,cascaded methods could be computationally intensive and the refined results are significantly dependent on the performance of its coarse segmentation results. To balance the segmentation accuracy and computational efficiency, we propose a Discriminative Feature Attention Network for pancreas segmentation, to effectively highlight pancreas features and improve segmentation accuracy without explicit pancreas location. The final segmentation is obtained by applying a simple yet effective post-processing step. Two experiments on both public NIH pancreas CT dataset and abdominal BTCV multi-organ dataset are individually conducted to show the effectiveness of our method for 2 D pancreas segmentation. We obtained average Dice Similarity Coefficient(DSC) of 82.82±6.09%, average Jaccard Index(JI) of 71.13± 8.30% and average Symmetric Average Surface Distance(ASD) of 1.69 ± 0.83 mm on the NIH dataset. Compared to the existing deep learning-based pancreas segmentation methods, our experimental results achieve the best average DSC and JI value.

Unsupervised Domain Adaptation Based on Discriminative Subspace Learning for Cross-Project Defect Prediction

:Cross-project defect prediction(CPDP)aims to predict the defects on target project by using a prediction model built on source projects.The main problem in CPDP is the huge distribution gap between the source project and the target project,which prevents the prediction model from performing well.Most existing methods overlook the class discrimination of the learned features.Seeking an effective transferable model from the source project to the target project for CPDP is challenging.In this paper,we propose an unsupervised domain adaptation based on the discriminative subspace learning(DSL)approach for CPDP.DSL treats the data from two projects as being from two domains and maps the data into a common feature space.It employs crossdomain alignment with discriminative information from different projects to reduce the distribution difference of the data between different projects and incorporates the class discriminative information.Specifically,DSL first utilizes subspace learning based domain adaptation to reduce the distribution gap of data between different projects.Then,it makes full use of the class label information of the source project and transfers the discrimination ability of the source project to the target project in the common space.Comprehensive experiments on five projects verify that DSL can build an effective prediction model and improve the performance over the related competing methods by at least 7.10%and 11.08%in terms of G-measure and AUC.

Balanced Discriminative Transfer Feature Learning for Visual Domain Adaptation

Transfer learning aims to transfer source models to a target domain.Leveraging the feature matching can alleviate the domain shift effectively,but this process ignores the relationship of the marginal distribution matching and the conditional distribution matching.Simultaneously,the discriminative information of both domains is also neglected,which is important for improving the performance on the target domain.In this paper,we propose a novel method called Balanced Discriminative Transfer Feature Learning for Visual Domain Adaptation(BDTFL).The proposed method can adaptively balance the relationship of both distribution matchings and capture the category discriminative information of both domains.Therefore,balanced feature matching can achieve more accurate feature matching and adaptively adjust itself to different scenes.At the same time,discriminative information is exploited to alleviate category confusion during feature matching.And with assistance of the category discriminative information captured from both domains,the source classifier can be transferred to the target domain more accurately and boost the performance of target classification.Extensive experiments show the superiority of BDTFL on popular visual cross-domain benchmarks.

Learning a discriminative high-fidelity dictionary for single channel source separation

Sparse-representation-based single-channel source separation,which aims to recover each source’s signal using its corresponding sub-dictionary,has attracted many scholars’attention.The basic premise of this model is that each sub-dictionary possesses discriminative information about its corresponding source,and this information can be used to recover almost every sample from that source.However,in a more general sense,the samples from a source are composed not only of discriminative information but also common information shared with other sources.This paper proposes learning a discriminative high-fidelity dictionary to improve the separation performance.The innovations are threefold.Firstly,an extra sub-dictionary was combined into a conventional union dictionary to ensure that the source-specific sub-dictionaries can capture only the purely discriminative information for their corresponding sources because the common information is collected in the additional sub-dictionary.Secondly,a task-driven learning algorithm is designed to optimize the new union dictionary and a set of weights that indicate how much of the common information should be allocated to each source.Thirdly,a source separation scheme based on the learned dictionary is presented.Experimental results on a human speech dataset yield evidence that our algorithm can achieve better separation performance than either state-of-the-art or traditional algorithms.

DISCRIMINATIVE ANALYSIS OF FUNCTIONAL NEAR-INFRARED SPECTROSCOPY SIGNALS FOR DEVELOPMENT OF NEUROIMAGING BIOMARKERS OF ELDERLY DEPRESSION

Functional near-infrared spectroscopy(fNIRS)is a neuroimaging technology which is suitable for psychiatric patients.Several fNIRS studies have found abnormal brain activations during cognitive tasks in elderly depression.In this paper,we proposed a discriminative model of multivariate pattern classification based on fNIRS signals to distinguish elderly depressed patients from healthy controls.This model used the brain activation patterns during a verbal fluency task as features of classification.Then Pseudo-Fisher Linear Discriminant Analysis was performed on the feature space to generate discriminative model.Using leave-one-out(LOO)cross-validation,our results showed a correct classification rate of 88%.The discriminative model showed its ability to identify people with elderly depression and suggested that fNIRS may be an efficient clinical tool for diagnosis of depression.This study may provide the first step for the development of neuroimaging biomarkers based on fNIRS in psychiatric disorders.

Discriminatively Constrained Semi-Supervised Multi-View Nonnegative Matrix Factorization with Graph Regularization

Nonnegative Matrix Factorization(NMF)is one of the most popular feature learning technologies in the field of machine learning and pattern recognition.It has been widely used and studied in the multi-view clustering tasks because of its effectiveness.This study proposes a general semi-supervised multi-view nonnegative matrix factorization algorithm.This algorithm incorporates discriminative and geometric information on data to learn a better-fused representation,and adopts a feature normalizing strategy to align the different views.Two specific implementations of this algorithm are developed to validate the effectiveness of the proposed framework:Graph regularization based Discriminatively Constrained Multi-View Nonnegative Matrix Factorization(GDCMVNMF)and Extended Multi-View Constrained Nonnegative Matrix Factorization(ExMVCNMF).The intrinsic connection between these two specific implementations is discussed,and the optimization based on multiply update rules is presented.Experiments on six datasets show that the effectiveness of GDCMVNMF and ExMVCNMF outperforms several representative unsupervised and semi-supervised multi-view NMF approaches.

Augmentation-based discriminative meta-learning for cross-machine few-shot fault diagnosis

Deep learning methods have demonstrated promising performance in fault diagnosis tasks.Although the scarcity of data in industrial scenarios limits the practical application of such methods,transfer learning effectively tackles this issue through crossmachine knowledge transfer.Nevertheless,the cross-machine few-shot problem,which is a more general industrial scenario,has been rarely investigated.Existing studies have not considered the cross-machine domain shift problem,which results in poor testing performance.This paper proposes an augmentation-based discriminative meta-learning method to address this issue.In the meta-training process,signal transformation is proposed to increase the meta-task diversity for more robust feature learning,and multi-scale learning is combined for more adaptive feature embedding.In the meta-testing process,limited labeled fault information is used to promote model generalization in the target domain through quasi-meta-training based on data augmentation.Furthermore,a novel hyperbolic prototypical loss is proposed for more discriminative feature representation and separable category prototypes by designing a hyperbolic decision boundary.Cross-machine few-shot diagnosis experiments were conducted using three datasets from different machines,namely,the bearing,motor,and gear datasets.The effectiveness of the proposed method was verified through ablation and comparison studies.

Underwater Pulse Waveform Recognition Based on Hash Aggregate Discriminant Network

Underwater pulse waveform recognition is an important method for underwater object detection.Most existing works focus on the application of traditional pattern recognition methods,which ignore the time-and space-varying characteristics in sound propagation channels and cannot easily extract valuable waveform features.Sound propagation channels in seawater are time-and space-varying convolutional channels.In the extraction of the waveform features of underwater acoustic signals,the effect of high-accuracy underwater acoustic signal recognition is identified by eliminating the influence of time-and space-varying convolutional channels to the greatest extent possible.We propose a hash aggregate discriminative network(HADN),which combines hash learning and deep learning to minimize the time-and space-varying effects on convolutional channels and adaptively learns effective underwater waveform features to achieve high-accuracy underwater pulse waveform recognition.In the extraction of the hash features of acoustic signals,a discrete constraint between clusters within a hash feature class is introduced.This constraint can ensure that the influence of convolutional channels on hash features is minimized.In addition,we design a new loss function called aggregate discriminative loss(AD-loss).The use of AD-loss and softmax-loss can increase the discriminativeness of the learned hash features.Experimental results show that on pool and ocean datasets,which were collected in pools and oceans,respectively,by using acoustic collectors,the proposed HADN performs better than other comparative models in terms of accuracy and mAP.

Robust Speech Recognition Method Based on Discriminative Environment Feature Extraction

It is an effective approach to learn the influence of environmental parameters, such as additive noise and channel distortions, from training data for robust speech recognition. Most of the previous methods are based on maximum likelihood estimation criterion. However, these methods do not lead to a minimum error rate result. In this paper, a novel discrimina-tive learning method of environmental parameters, which is based on Minimum Classification Error (MCE) criterion, is proposed. In the method, a simple classifier and the Generalized Probabilistic Descent (GPD) algorithm are adopted to iteratively learn the environmental pa-rameters. Consequently, the clean speech features are estimated from the noisy speech features with the estimated environmental parameters, and then the estimations of clean speech features are utilized in the back-end HMM classifier. Experiments show that the best error rate reduction of 32.1% is obtained, tested on a task of 18 isolated confusion Korean words, relative to a conventional HMM system.

Experimental Study of Discriminative Adaptive Training and MLLR for Automatic Pronunciation Evaluation

A stronger canonical model was developed to improve the performance of automatic pronunciation evaluations. Three different strategies were investigated with speaker adaptive training to normalize variations among speakers, minimum phone error training to identify easily confused phones and maximum likelihood linear regression （MLLR） adaptation to compensate for accent variations between native and non-native speakers. The three schemes were combined to improve the correlation coefficient between machine scores and human scores from 0.651 to 0.679 on the sentence level and from 0.788 to 0.822 on the speaker level.

Discriminative tonal feature extraction method in mandarin speech recognition

To utilize the supra-segmental nature of Mandarin tones, this article proposes a feature extraction method for hidden markov model （HMM） based tone modeling. The method uses linear transforms to project Fo（fundamental frequency） features of neighboring syllables as compensations, and adds them to the original Fo features of the current syUable. The transforms are discriminatively trained by using an objective function termed as ＂minimum tone error＂, which is a smooth approximation of tone recognition accuracy. Experiments show that the new tonal features achieve 3.82% tone recognition rate improvement, compared with the baseline, using maximum likelihood trained HMM on the normal F0 features. Further experiments show that discriminative HMM training on the new features is 8.78% better than the baseline.

A separation method of singing and accompaniment combining discriminative training deep neural network

For the difficulty of separation between singing and accompaniment in the musical signals,an improved music separation method of based on discriminative training depth neural network(DNN) was proposed.Firstly,based on the DNN model,considering the reconstruction errors and discrimination information between singing and accompaniment,an improved objective function was presented to discriminate the training;Then,an additional layer was added to DNN model,introducing the time-frequency masking to optimize the estimated accompaniment of the song,and the corresponding time-domain signal was obtained by inverse Fourier transform;Finally,the influence of different parameters on the separation performance was verified,and compared it with the existing music separation methods.The experimental results showed that the improved objective function and the introduction of time-frequency masking significantly improved the separation performance of the DNN,and the separation performance was improved about 4 dB compared with other existing music separation methods,thus verifying that the proposed method was an effective music separation algorithm.

LociScan,a tool for screening genetic marker combinations for plant variety discrimination

To reduce the cost and increase the efficiency of plant genetic marker fingerprinting for variety discrimination,it is desirable to identify the optimal marker combinations.We describe a marker combination screening model based on the genetic algorithm(GA)and implemented in a software tool,Loci Scan.Ratio-based variety discrimination power provided the largest optimization space among multiple fitness functions.Among GA parameters,an increase in population size and generation number enlarged optimization depth but also calculation workload.Exhaustive algorithm afforded the same optimization depth as GA but vastly increased calculation time.In comparison with two other software tools,Loci Scan accommodated missing data,reduced calculation time,and offered more fitness functions.In large datasets,the sample size of training data exerted the strongest influence on calculation time,whereas the marker size of training data showed no effect,and target marker number had limited effect on analysis speed.

Impact pressure of waves generated by landslides on bank slopes

Impulse waves that are generated by landslides in narrow reservoir areas threaten the stability of buildings and bank slopes.To discuss the action process and evolution law of the wave pressure on bank slopes,a three-dimensional physical model test that considers impulse waves generated by landslides was performed,and factors including landslide width,thickness,slope angles of the sliding surface,and bank slope angle were considered.Based on wave forms on the bank slopes,wave pressure curve characteristics,and peak value,the action process of wave pressure could be divided into the following stages:maximum pulsating pressure stage,wave impact stage(when waves break),and stationary pulsation stage.It was found that wave breaking is dependent on the value of the surf similarity parameterξ.The distribution pattern of impact pressure decays linearly on both sides of the maximum impact pressure point,and the attenuation degree decreases when it attains 40%of the maximum value.Thus,it is proposed that the prediction formula for the maximum effective impact pressure of the bank slope be related to the reciprocal of wave steepness,relative water depth,and slope rate.The prediction formula provides strong theoretical support for early safety warning and for predicting the bank slope under impulse waves generated by landslides.

The Analysis of the Correlation between SPT and CPT Based on CNN-GA and Liquefaction Discrimination Research

The objective of this study is to investigate themethods for soil liquefaction discrimination. Typically, predicting soilliquefaction potential involves conducting the standard penetration test (SPT), which requires field testing and canbe time-consuming and labor-intensive. In contrast, the cone penetration test (CPT) provides a more convenientmethod and offers detailed and continuous information about soil layers. In this study, the feature matrix based onCPT data is proposed to predict the standard penetration test blow count N. The featurematrix comprises the CPTcharacteristic parameters at specific depths, such as tip resistance qc, sleeve resistance f s, and depth H. To fuse thefeatures on the matrix, the convolutional neural network (CNN) is employed for feature extraction. Additionally,Genetic Algorithm (GA) is utilized to obtain the best combination of convolutional kernels and the number ofneurons. The study evaluated the robustness of the proposed model using multiple engineering field data sets.Results demonstrated that the proposed model outperformed conventional methods in predicting N values forvarious soil categories, including sandy silt, silty sand, and clayey silt. Finally, the proposed model was employedfor liquefaction discrimination. The liquefaction discrimination based on the predicted N values was comparedwith the measured N values, and the results showed that the discrimination results were in 75% agreement. Thestudy has important practical application value for foundation liquefaction engineering. Also, the novel methodadopted in this research provides new ideas and methods for research in related fields, which is of great academicsignificance.

I-DCGAN and TOPSIS-IFP:A simulation generation model for radiographic flaw detection images in light alloy castings and an algorithm for quality evaluation of generated images

The intelligent detection technology driven by X-ray images and deep learning represents the forefront of advanced techniques and development trends in flaw detection and automated evaluation of light alloy castings.However,the efficacy of deep learning models hinges upon a substantial abundance of flaw samples.The existing research on X-ray image augmentation for flaw detection suffers from shortcomings such as poor diversity of flaw samples and low reliability of quality evaluation.To this end,a novel approach was put forward,which involves the creation of the Interpolation-Deep Convolutional Generative Adversarial Network(I-DCGAN)for flaw detection image generation and a comprehensive evaluation algorithm named TOPSIS-IFP.I-DCGAN enables the generation of high-resolution,diverse simulated images with multiple appearances,achieving an improvement in sample diversity and quality while maintaining a relatively lower computational complexity.TOPSIS-IFP facilitates multi-dimensional quality evaluation,including aspects such as diversity,authenticity,image distribution difference,and image distortion degree.The results indicate that the X-ray radiographic images of magnesium and aluminum alloy castings achieve optimal performance when trained up to the 800th and 600th epochs,respectively.The TOPSIS-IFP value reaches 78.7%and 73.8%similarity to the ideal solution,respectively.Compared to single index evaluation,the TOPSIS-IFP algorithm achieves higher-quality simulated images at the optimal training epoch.This approach successfully mitigates the issue of unreliable quality associated with single index evaluation.The image generation and comprehensive quality evaluation method developed in this paper provides a novel approach for image augmentation in flaw recognition,holding significant importance for enhancing the robustness of subsequent flaw recognition networks.