With the rapid growth of power systems measurements in terms of size and complexity,discovering statistical patterns for a large variety of real-world applications such as renewable energy prediction,demand response,e...With the rapid growth of power systems measurements in terms of size and complexity,discovering statistical patterns for a large variety of real-world applications such as renewable energy prediction,demand response,energy disaggregation,and state estimation is considered a crucial challenge.In recent years,deep learning has emerged as a novel class of machine learning algorithms that represents power systems data via a large hypothesis space that leads to the state-of-the-art performance compared to most recent data-driven algorithms.This study explores the theoretical advantages of deep representation learning in power systems research.We review deep learning methodologies presented and applied in a wide range of supervised,unsupervised,and semi-supervised applications as well as reinforcement learning tasks.We discuss various settings of problems solved by discriminative deep models including stacked autoencoders and convolutional neural networks as well as generative deep architectures such as deep belief networks and variational autoencoders.The theoretical and experimental analysis of deep neural networks in this study motivates longterm research on optimizing this cutting-edge class of models to achieve significant improvements in the future power systems research.展开更多
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 t...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.展开更多
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 in...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.展开更多
Objective:Amphenicols(chloramphenicol,thiamphenicol and forfenicol)can cause aplastic anaemia and other severe side effects to consumers;therefore,it is necessary to inspect their residues in foods of animal origin.Ho...Objective:Amphenicols(chloramphenicol,thiamphenicol and forfenicol)can cause aplastic anaemia and other severe side effects to consumers;therefore,it is necessary to inspect their residues in foods of animal origin.However,there has been no report on the use of amphenicols receptor for the determination of their residues,and none of the previously reported immunoassays for amphenicols can differentiate the specifc species.Materials and Methods:In this study,the ribosomal protein L16 of Escherichia coli was frst expressed,and its intermolecular interaction mechanisms with the three amphenicols was studied using the molecular docking technique.The protein was then combined with three enzymelabelled conjugates to develop a direct competitive array on microplate for determination of the three drugs in egg.Results:Due to the use of principal component analysis to analyse the data,this method could discriminate the three drugs in the range 0.1–10 ng/mL,and the limits of detection for the three drugs were in the range of 0.0002–0.0009 ng/mL.The analysis results for the unknown egg samples were consistent with a liquid chromatography–tandem mass spectrometry method,and the method performances were superior to the previous immunoassays for amphenicols.Conclusion:This is the frst paper reporting the use of ribosomal protein L16 to develop a competitive array for discriminative determination of amphenicols in food samples.展开更多
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-vary...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.展开更多
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 spe...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.展开更多
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 li...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.展开更多
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 effective...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.展开更多
An important component of a spoken term detection （STD） system involves estimating confidence measures of hypothesised detections.A potential problem of the widely used lattice-based confidence estimation,however,is...An important component of a spoken term detection （STD） system involves estimating confidence measures of hypothesised detections.A potential problem of the widely used lattice-based confidence estimation,however,is that the confidence scores are treated uniformly for all search terms,regardless of how much they may differ in terms of phonetic or linguistic properties.This problem is particularly evident for out-of-vocabulary （OOV） terms which tend to exhibit high intra-term diversity.To address the impact of term diversity on confidence measures,we propose in this work a term-dependent normalisation technique which compensates for term diversity in confidence estimation.We first derive an evaluation-metric-oriented normalisation that optimises the evaluation metric by compensating for the diverse occurrence rates among terms,and then propose a linear bias compensation and a discriminative compensation to deal with the bias problem that is inherent in lattice-based confidence measurement and from which the Term Specific Threshold （TST） approach suffers.We tested the proposed technique on speech data from the multi-party meeting domain with two state-ofthe-art STD systems based on phonemes and words respectively.The experimental results demonstrate that the confidence normalisation approach leads to a significant performance improvement in STD,particularly for OOV terms with phonemebased systems.展开更多
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 ...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.展开更多
A robust and eficient feature matching method is necessary for visual navigation in asteroid-landing missions.Based on the visual navigation framework and motion characteristics of asteroids,a robust and efficient tem...A robust and eficient feature matching method is necessary for visual navigation in asteroid-landing missions.Based on the visual navigation framework and motion characteristics of asteroids,a robust and efficient template feature matching method is proposed to adapt to feature distortion and scale change cases for visual navigation of asteroids.The proposed method is primarily based on a motion-constrained discriminative correlation filter(DCF).The prior information provided by the motion constraints between sequence images is used to provide a predicted search region for template feature matching.Additionally,some specific template feature samples are generated using the motion constraints for correlation filter learning,which is beneficial for training a scale and feature distortion adaptive correlation filter for accurate feature matching.Moreover,average peak-to-correlation energy(APCE)and jointly consistent measurements(JCMs)were used to eliminate false matching.Images captured by the Touch And Go Camera System(TAGCAMS)of the Bennu asteroid were used to evaluate the performance of the proposed method.In particular,both the robustness and accuracy of region matching and template center matching are evaluated.The qualitative and quantitative results illustrate the advancement of the proposed method in adapting to feature distortions and large-scale changes during spacecraft landing.展开更多
Recently, data-driven methods, especially deep learning, outperform other methods for rolling elementbearing (REB) fault diagnosis. Nevertheless, most research work assumes that REB dataset is unbiased. Inthe real ind...Recently, data-driven methods, especially deep learning, outperform other methods for rolling elementbearing (REB) fault diagnosis. Nevertheless, most research work assumes that REB dataset is unbiased. Inthe real industry applications, the dataset bias exists with REB owing to varying REB working conditions andnoise interference. Recently proposed adversarial discriminative domain adaptation (ADDA) is an increasinglypopular incarnation to solve dataset bias problem. However, it mainly devotes to realizing domain alignments, andignores class-level alignments;it can cause degradation of classification performance. In this study, we proposea new REB fault diagnosis model based on improved ADDA to address dataset bias. The proposed diagnosismodel realizes domain- and class-level alignments in dataset bias scenario;it consists of two feature extractors,a domain discriminator, and two label classifiers. The feature extractors and domain discriminator are trainedin an adversarial manner to minimize the domain difference in feature extractors. The domain discrepancy inlabel classifier is reduced by minimizing correlation alignment (CORAL) loss. We evaluate the proposed model onthe Case Western Reserve University (CWRU) bearing dataset and Paderborn University bearing dataset. Theproposed method yields better results than other methods and has good prospects for industrial applications.展开更多
Background:As the demand for traditional Chinese medicinal materials increases in China and even the world,there is an urgent need for an effective and simple identification technology to identify the origin and quali...Background:As the demand for traditional Chinese medicinal materials increases in China and even the world,there is an urgent need for an effective and simple identification technology to identify the origin and quality of the latter and ensure the safety of clinical medication.Mineral element analysis and isotope finger-printing are the two commonly used techniques in traditional origin identification.Both of these techniques require the use of stoichiometric methods in the identification process.Although they have high accuracy and sensitivity,they are expensive and inefficient.In addition,near-infrared spectroscopy is a fast,nondestructive,and widely used identification technique developed in recent years,but its identification results are susceptible to samples’states and environmental conditions,and its sensitivity is low.Hyperspectral imaging combines the advantages of imaging technology and optical technology,which can simultaneously access the image information and spectral information which reflect the external characteristics,internal physical structure,and chemical composition of the samples.Hyperspectral imaging is widely applied to agricultural product inspection,but research into its application in origin and quality identification of TCM materials is rare.Methods:In this study,the algorithm framework discriminative marginalized least squares regression(DMLSR)was used for feature extraction of frankincense hyperspectral data.The DMLSR with intraclass compactness graph and manifold regularization can efficiently learn the projective samples with higher separability and less redundant information than the original samples.Then,the discriminative collaborative representation with Tikhonov regularization(DCRT)was applied for classifying the geographical origin and level of frankincense.DCRT introduces the discriminant regularization term and incorporates SID,which is more sensitive to the spectrum as the measurement method and is more suitable for the frankincense spectral data compared with SVM.Resul展开更多
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（fundamen...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.展开更多
基金supported by the Science and Technology Project of State Grid Corporation of China(No.5455HJ180018).
文摘With the rapid growth of power systems measurements in terms of size and complexity,discovering statistical patterns for a large variety of real-world applications such as renewable energy prediction,demand response,energy disaggregation,and state estimation is considered a crucial challenge.In recent years,deep learning has emerged as a novel class of machine learning algorithms that represents power systems data via a large hypothesis space that leads to the state-of-the-art performance compared to most recent data-driven algorithms.This study explores the theoretical advantages of deep representation learning in power systems research.We review deep learning methodologies presented and applied in a wide range of supervised,unsupervised,and semi-supervised applications as well as reinforcement learning tasks.We discuss various settings of problems solved by discriminative deep models including stacked autoencoders and convolutional neural networks as well as generative deep architectures such as deep belief networks and variational autoencoders.The theoretical and experimental analysis of deep neural networks in this study motivates longterm research on optimizing this cutting-edge class of models to achieve significant improvements in the future power systems research.
基金This work was supported by the National Key Research and Development Project of China(No.2019YFB2102500)the Strategic Priority CAS Project(No.XDB38040200)+2 种基金the National Natural Science Foundation of China(Nos.62206269,U1913210)the Guangdong Provincial Science and Technology Projects(Nos.2022A1515011217,2022A1515011557)the Shenzhen Science and Technology Projects(No.JSGG20211029095546003)。
文摘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.
基金the Natural Science Foundation of Henan Province(232300420094)the Science and TechnologyResearch Project of Henan Province(222102220092).
文摘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.
基金the National Natural Science Foundation of China(No.32372447)the Natural Science Foundation of Hebei Province(No.C2023204045),China.
文摘Objective:Amphenicols(chloramphenicol,thiamphenicol and forfenicol)can cause aplastic anaemia and other severe side effects to consumers;therefore,it is necessary to inspect their residues in foods of animal origin.However,there has been no report on the use of amphenicols receptor for the determination of their residues,and none of the previously reported immunoassays for amphenicols can differentiate the specifc species.Materials and Methods:In this study,the ribosomal protein L16 of Escherichia coli was frst expressed,and its intermolecular interaction mechanisms with the three amphenicols was studied using the molecular docking technique.The protein was then combined with three enzymelabelled conjugates to develop a direct competitive array on microplate for determination of the three drugs in egg.Results:Due to the use of principal component analysis to analyse the data,this method could discriminate the three drugs in the range 0.1–10 ng/mL,and the limits of detection for the three drugs were in the range of 0.0002–0.0009 ng/mL.The analysis results for the unknown egg samples were consistent with a liquid chromatography–tandem mass spectrometry method,and the method performances were superior to the previous immunoassays for amphenicols.Conclusion:This is the frst paper reporting the use of ribosomal protein L16 to develop a competitive array for discriminative determination of amphenicols in food samples.
基金partially supported by the National Key Research and Development Program of China(No.2018 AAA0100400)the Natural Science Foundation of Shandong Province(Nos.ZR2020MF131 and ZR2021ZD19)the Science and Technology Program of Qingdao(No.21-1-4-ny-19-nsh).
文摘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.
基金Supported by the National High-Tech Research and Development (863) Program of China (No. 2008AA01Z118)
文摘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.
基金The authors would like to acknowledge the financial support from the National Natural Science Foundation of China(No.61379145)the Joint Funds of CETC(Grant No.20166141B08020101).
文摘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.
基金supported by the National Natural Science Foundation of China(Grant No.51975356)Shanghai Municipal Science and Technology Major Project(Grant No.2021SHZDZX0102)。
文摘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.
文摘An important component of a spoken term detection （STD） system involves estimating confidence measures of hypothesised detections.A potential problem of the widely used lattice-based confidence estimation,however,is that the confidence scores are treated uniformly for all search terms,regardless of how much they may differ in terms of phonetic or linguistic properties.This problem is particularly evident for out-of-vocabulary （OOV） terms which tend to exhibit high intra-term diversity.To address the impact of term diversity on confidence measures,we propose in this work a term-dependent normalisation technique which compensates for term diversity in confidence estimation.We first derive an evaluation-metric-oriented normalisation that optimises the evaluation metric by compensating for the diverse occurrence rates among terms,and then propose a linear bias compensation and a discriminative compensation to deal with the bias problem that is inherent in lattice-based confidence measurement and from which the Term Specific Threshold （TST） approach suffers.We tested the proposed technique on speech data from the multi-party meeting domain with two state-ofthe-art STD systems based on phonemes and words respectively.The experimental results demonstrate that the confidence normalisation approach leads to a significant performance improvement in STD,particularly for OOV terms with phonemebased systems.
基金the '863' High-Tech Programme of China (No. 863-306ZT03-02-3) and partially by the National Natural Science Foundation of China
文摘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.
基金funded by the National Natural Science Foundation of China under Grant Nos.41822106 and 42101447the Dawn Scholar of Shanghai Program under Grant No.18SG22+2 种基金the Science and Technology on Aerospace Flight Dynamics Laboratory,China,under Grant No.KGJ6142210110305State Key Laboratory of Disaster Reduction in Civil Engineering under Grant No.SLDRCE19-B-35Fundamental Research Funds for the Central Universities of China.
文摘A robust and eficient feature matching method is necessary for visual navigation in asteroid-landing missions.Based on the visual navigation framework and motion characteristics of asteroids,a robust and efficient template feature matching method is proposed to adapt to feature distortion and scale change cases for visual navigation of asteroids.The proposed method is primarily based on a motion-constrained discriminative correlation filter(DCF).The prior information provided by the motion constraints between sequence images is used to provide a predicted search region for template feature matching.Additionally,some specific template feature samples are generated using the motion constraints for correlation filter learning,which is beneficial for training a scale and feature distortion adaptive correlation filter for accurate feature matching.Moreover,average peak-to-correlation energy(APCE)and jointly consistent measurements(JCMs)were used to eliminate false matching.Images captured by the Touch And Go Camera System(TAGCAMS)of the Bennu asteroid were used to evaluate the performance of the proposed method.In particular,both the robustness and accuracy of region matching and template center matching are evaluated.The qualitative and quantitative results illustrate the advancement of the proposed method in adapting to feature distortions and large-scale changes during spacecraft landing.
基金Foundation item:the Research on Intelligent Ship Testing and Verification(No.[2018]473)。
文摘Recently, data-driven methods, especially deep learning, outperform other methods for rolling elementbearing (REB) fault diagnosis. Nevertheless, most research work assumes that REB dataset is unbiased. Inthe real industry applications, the dataset bias exists with REB owing to varying REB working conditions andnoise interference. Recently proposed adversarial discriminative domain adaptation (ADDA) is an increasinglypopular incarnation to solve dataset bias problem. However, it mainly devotes to realizing domain alignments, andignores class-level alignments;it can cause degradation of classification performance. In this study, we proposea new REB fault diagnosis model based on improved ADDA to address dataset bias. The proposed diagnosismodel realizes domain- and class-level alignments in dataset bias scenario;it consists of two feature extractors,a domain discriminator, and two label classifiers. The feature extractors and domain discriminator are trainedin an adversarial manner to minimize the domain difference in feature extractors. The domain discrepancy inlabel classifier is reduced by minimizing correlation alignment (CORAL) loss. We evaluate the proposed model onthe Case Western Reserve University (CWRU) bearing dataset and Paderborn University bearing dataset. Theproposed method yields better results than other methods and has good prospects for industrial applications.
文摘Background:As the demand for traditional Chinese medicinal materials increases in China and even the world,there is an urgent need for an effective and simple identification technology to identify the origin and quality of the latter and ensure the safety of clinical medication.Mineral element analysis and isotope finger-printing are the two commonly used techniques in traditional origin identification.Both of these techniques require the use of stoichiometric methods in the identification process.Although they have high accuracy and sensitivity,they are expensive and inefficient.In addition,near-infrared spectroscopy is a fast,nondestructive,and widely used identification technique developed in recent years,but its identification results are susceptible to samples’states and environmental conditions,and its sensitivity is low.Hyperspectral imaging combines the advantages of imaging technology and optical technology,which can simultaneously access the image information and spectral information which reflect the external characteristics,internal physical structure,and chemical composition of the samples.Hyperspectral imaging is widely applied to agricultural product inspection,but research into its application in origin and quality identification of TCM materials is rare.Methods:In this study,the algorithm framework discriminative marginalized least squares regression(DMLSR)was used for feature extraction of frankincense hyperspectral data.The DMLSR with intraclass compactness graph and manifold regularization can efficiently learn the projective samples with higher separability and less redundant information than the original samples.Then,the discriminative collaborative representation with Tikhonov regularization(DCRT)was applied for classifying the geographical origin and level of frankincense.DCRT introduces the discriminant regularization term and incorporates SID,which is more sensitive to the spectrum as the measurement method and is more suitable for the frankincense spectral data compared with SVM.Resul
文摘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.
