Domain adaptation aims to reduce the distribution gap between the training data(source domain)and the target data.This enables effective predictions even for domains not seen during training.However,most conventional ...Domain adaptation aims to reduce the distribution gap between the training data(source domain)and the target data.This enables effective predictions even for domains not seen during training.However,most conventional domain adaptation methods assume a single source domain,making them less suitable for modern deep learning settings that rely on diverse and large-scale datasets.To address this limitation,recent research has focused on Multi-Source Domain Adaptation(MSDA),which aims to learn effectively from multiple source domains.In this paper,we propose Efficient Domain Transition for Multi-source(EDTM),a novel and efficient framework designed to tackle two major challenges in existing MSDA approaches:(1)integrating knowledge across different source domains and(2)aligning label distributions between source and target domains.EDTM leverages an ensemble-based classifier expert mechanism to enhance the contribution of source domains that are more similar to the target domain.To further stabilize the learning process and improve performance,we incorporate imitation learning into the training of the target model.In addition,Maximum Classifier Discrepancy(MCD)is employed to align class-wise label distributions between the source and target domains.Experiments were conducted using Digits-Five,one of the most representative benchmark datasets for MSDA.The results show that EDTM consistently outperforms existing methods in terms of average classification accuracy.Notably,EDTM achieved significantly higher performance on target domains such as Modified National Institute of Standards and Technolog with blended background images(MNIST-M)and Street View House Numbers(SVHN)datasets,demonstrating enhanced generalization compared to baseline approaches.Furthermore,an ablation study analyzing the contribution of each loss component validated the effectiveness of the framework,highlighting the importance of each module in achieving optimal performance.展开更多
Automatically recognizing radar emitters from com-plex electromagnetic environments is important but non-trivial.Moreover,the changing electromagnetic environment results in inconsistent signal distribution in the rea...Automatically recognizing radar emitters from com-plex electromagnetic environments is important but non-trivial.Moreover,the changing electromagnetic environment results in inconsistent signal distribution in the real world,which makes the existing approaches perform poorly for recognition tasks in different scenes.In this paper,we propose a domain generaliza-tion framework is proposed to improve the adaptability of radar emitter signal recognition in changing environments.Specifically,we propose an end-to-end denoising based domain-invariant radar emitter recognition network(DDIRNet)consisting of a denoising model and a domain invariant representation learning model(IRLM),which mutually benefit from each other.For the signal denoising model,a loss function is proposed to match the feature of the radar signals and guarantee the effectiveness of the model.For the domain invariant representation learning model,contrastive learning is introduced to learn the cross-domain feature by aligning the source and unseen domain distri-bution.Moreover,we design a data augmentation method that improves the diversity of signal data for training.Extensive experiments on classification have shown that DDIRNet achieves up to 6.4%improvement compared with the state-of-the-art radar emitter recognition methods.The proposed method pro-vides a promising direction to solve the radar emitter signal recognition problem.展开更多
In the realm of medical image segmentation,particularly in cardiac magnetic resonance imaging(MRI),achieving robust performance with limited annotated data is a significant challenge.Performance often degrades when fa...In the realm of medical image segmentation,particularly in cardiac magnetic resonance imaging(MRI),achieving robust performance with limited annotated data is a significant challenge.Performance often degrades when faced with testing scenarios from unknown domains.To address this problem,this paper proposes a novel semi-supervised approach for cardiac magnetic resonance image segmentation,aiming to enhance predictive capabilities and domain generalization(DG).This paper establishes an MT-like model utilizing pseudo-labeling and consistency regularization from semi-supervised learning,and integrates uncertainty estimation to improve the accuracy of pseudo-labels.Additionally,to tackle the challenge of domain generalization,a data manipulation strategy is introduced,extracting spatial and content-related information from images across different domains,enriching the dataset with a multi-domain perspective.This papers method is meticulously evaluated on the publicly available cardiac magnetic resonance imaging dataset M&Ms,validating its effectiveness.Comparative analyses against various methods highlight the out-standing performance of this papers approach,demonstrating its capability to segment cardiac magnetic resonance images in previously unseen domains even with limited annotated data.展开更多
In actual industrial scenarios,the variation of operating conditions,the existence of data noise,and failure of measurement equipment will inevitably affect the distribution of perceptive data.Deep learning-based faul...In actual industrial scenarios,the variation of operating conditions,the existence of data noise,and failure of measurement equipment will inevitably affect the distribution of perceptive data.Deep learning-based fault diagnosis algorithms strongly rely on the assumption that source and target data are independent and identically distributed,and the learned diagnosis knowledge is difficult to generalize to out-of-distribution data.Domain generalization(DG)aims to achieve the generalization of arbitrary target domain data by using only limited source domain data for diagnosis model training.The research of DG for fault diagnosis has made remarkable progress in recent years and lots of achievements have been obtained.In this article,for the first time a comprehensive literature review on DG for fault diagnosis from a learning mechanism-oriented perspective is provided to summarize the development in recent years.Specifically,we first conduct a comprehensive review on existing methods based on the similarity of basic principles and design motivations.Then,the recent trend of DG for fault diagnosis is also analyzed.Finally,the existing problems and future prospect is performed.展开更多
Single-source Domain Generalization(SDG)is a promising yet challenging technology that aims to transfer knowledge from a singular source domain to multiple and unfamiliar target domains.Existing SDG methods typically ...Single-source Domain Generalization(SDG)is a promising yet challenging technology that aims to transfer knowledge from a singular source domain to multiple and unfamiliar target domains.Existing SDG methods typically rely on domain expansion to implement data variation and broaden the coverage of the training domain.However,due to the lack of proper semantic consistency and sample diversity constraints,these methods have limited improvement in generalization performance for most practical applications.In this paper,we propose a Causality-Aware Single-source Domain Generalization(CASDG)method to utilize both semantic consistency and diversity during the data transformation process.First,a causality-aware module is designed to accurately measure the causal effect between latent features and labels.Then,we introduce a causal domain expansion module,which utilizes the causal effect matrix as a semantic consistency constraint and mutual information as a sample diversity constraint.These two constraints are jointly used to encourage the style transformer to generate new auxiliary samples that are undeviated from the original samples.The image classification model using our method can produce the best classification performance for unknown domain data compared to the state-of-the-art methods.展开更多
Domain randomization is a widely adopted technique in deep reinforcement learning(DRL)to improve agent generalization by exposing policies to diverse environmental conditions.This paper investigates the impact of diff...Domain randomization is a widely adopted technique in deep reinforcement learning(DRL)to improve agent generalization by exposing policies to diverse environmental conditions.This paper investigates the impact of different reset strategies,normal,non-randomized,and randomized,on agent performance using the Deep Deterministic Policy Gradient(DDPG)and Twin Delayed DDPG(TD3)algorithms within the CarRacing-v2 environment.Two experimental setups were conducted:an extended training regime with DDPG for 1000 steps per episode across 1000 episodes,and a fast execution setup comparing DDPG and TD3 for 30 episodes with 50 steps per episode under constrained computational resources.A step-based reward scaling mechanism was applied under the randomized reset condition to promote broader state exploration.Experimental results showthat randomized resets significantly enhance learning efficiency and generalization,with DDPG demonstrating superior performance across all reset strategies.In particular,DDPG combined with randomized resets achieves the highest smoothed rewards(reaching approximately 15),best stability,and fastest convergence.These differences are statistically significant,as confirmed by t-tests:DDPG outperforms TD3 under randomized(t=−101.91,p<0.0001),normal(t=−21.59,p<0.0001),and non-randomized(t=−62.46,p<0.0001)reset conditions.The findings underscore the critical role of reset strategy and reward shaping in enhancing the robustness and adaptability of DRL agents in continuous control tasks,particularly in environments where computational efficiency and training stability are crucial.展开更多
Learning domain-invariant feature representations is critical to alleviate the distribution differences between training and testing domains.The existing mainstream domain generalization approaches primarily pursue to...Learning domain-invariant feature representations is critical to alleviate the distribution differences between training and testing domains.The existing mainstream domain generalization approaches primarily pursue to align the across-domain distributions to extract the transferable feature representations.However,these representations may be insufficient and unstable.Moreover,these networks may also undergo catastrophic forgetting because the previous learned knowledge is replaced by the new learned knowledge.To cope with these issues,we propose a novel causality-based contrastive incremental learning model for domain generalization,which mainly includes three components:(1)intra-domain causal factorization,(2)inter-domain Mahalanobis similarity metric,and(3)contrastive knowledge distillation.The model extracts intra and inter domain-invariant knowledge to improve model generalization.Specifically,we first introduce a causal factorization to extract intra-domain invariant knowledge.Then,we design a Mahalanobis similarity metric to extract common inter-domain invariant knowledge.Finally,we propose a contrastive knowledge distillation with exponential moving average to distill model parameters in a smooth way to preserve the previous learned knowledge and mitigate model forgetting.Extensive experiments on several domain generalization benchmarks prove that our model achieves the state-of-the-art results,which sufficiently show the effectiveness of our model.展开更多
We dealt in a series of previous publications with some geometric aspects of the mappings by functions obtained as analytic continuations to the whole complex plane of general Dirichlet series. Pictures illustrating t...We dealt in a series of previous publications with some geometric aspects of the mappings by functions obtained as analytic continuations to the whole complex plane of general Dirichlet series. Pictures illustrating those aspects contain a lot of other information which has been waiting for a rigorous proof. Such a task is partially fulfilled in this paper, where we succeeded among other things, to prove a theorem about general Dirichlet series having as corollary the Speiser’s theorem. We have also proved that those functions do not possess multiple zeros of order higher than 2 and the double zeros have very particular locations. Moreover, their derivatives have only simple zeros. With these results at hand, we revisited GRH for a simplified proof.展开更多
Smart manufacturing suffers from the heterogeneity of local data distribution across parties,mutual information silos and lack of privacy protection in the process of industry chain collaboration.To address these prob...Smart manufacturing suffers from the heterogeneity of local data distribution across parties,mutual information silos and lack of privacy protection in the process of industry chain collaboration.To address these problems,we propose a federated domain adaptation algorithm based on knowledge distillation and contrastive learning.Knowledge distillation is used to extract transferable integration knowledge from the different source domains and the quality of the extracted integration knowledge is used to assign reasonable weights to each source domain.A more rational weighted average aggregation is used in the aggregation phase of the center server to optimize the global model,while the local model of the source domain is trained with the help of contrastive learning to constrain the local model optimum towards the global model optimum,mitigating the inherent heterogeneity between local data.Our experiments are conducted on the largest domain adaptation dataset,and the results show that compared with other traditional federated domain adaptation algorithms,the algorithm we proposed trains a more accurate model,requires fewer communication rounds,makes more effective use of imbalanced data in the industrial area,and protects data privacy.展开更多
Previous studies have already shown that Raman spectroscopy can be used in the encoding of suspension array technology.However,almost all existing convolutional neural network-based decoding approaches rely on supervi...Previous studies have already shown that Raman spectroscopy can be used in the encoding of suspension array technology.However,almost all existing convolutional neural network-based decoding approaches rely on supervision with ground truth,and may not be well generalized to unseen datasets,which were collected under different experimental conditions,applying with the same coded material.In this study,we propose an improved model based on CyCADA,named as Detail constraint Cycle Domain Adaptive Model(DCDA).DCDA implements the clasification of unseen datasets through domain adaptation,adapts representations at the encode level with decoder-share,and enforces coding features while leveraging a feat loss.To improve detailed structural constraints,DCDA takes downsample connection and skips connection.Our model improves the poor generalization of existing models and saves the cost of the labeling process for unseen target datasets.Compared with other models,extensive experiments and ablation studies show the superiority of DCDA in terms of classification stability and generalization.The model proposed by the research achieves a classification with an accuracy of 100%when applied in datasets,in which the spectrum in the source domain is far less than the target domain.展开更多
In this paper,we first obtain a unified integral representation on the analytic varieties of the general bounded domain in Stein manifolds(the two types bounded domains in[3]are regarded as its special cases).Secondly...In this paper,we first obtain a unified integral representation on the analytic varieties of the general bounded domain in Stein manifolds(the two types bounded domains in[3]are regarded as its special cases).Secondly we get the integral formulas of the solution of∂-equation.And we use a new and unique method to give a uniform estimate of the solution of∂-equation,which is different from Henkin's method.展开更多
Analysis of functional MRI (fMRI) blood oxygenation level dependent (BOLD) data is typically carried out in the time domain where the data has a high temporal correlation. These analyses usually employ parametric mode...Analysis of functional MRI (fMRI) blood oxygenation level dependent (BOLD) data is typically carried out in the time domain where the data has a high temporal correlation. These analyses usually employ parametric models of the hemodynamic response function (HRF) where either pre-whitening of the data is attempted or autoregressive (AR) models are employed to model the noise. Statistical analysis then proceeds via regression of the convolution of the HRF with the input stimuli. This approach has limitations when considering that the time series collected are embedded in a brain image in which the AR model order may vary and pre-whitening techniques may be insufficient for handling faster sampling times. However fMRI data can be analyzed in the Fourier domain where the assumptions made as to the structure of the noise can be less restrictive and hypothesis tests are straightforward for single subject analysis, especially useful in a clinical setting. This allows for experiments that can have both fast temporal sampling and event-related designs where stimuli can be closely spaced in time. Equally important, statistical analysis in the Fourier domain focuses on hypothesis tests based on nonparametric estimates of the hemodynamic transfer function (HRF in the frequency domain). This is especially important for experimental designs involving multiple states (drug or stimulus induced) that may alter the form of the response function. In this context a univariate general linear model in the Fourier domain has been applied to analyze BOLD data sampled at a rate of 400 ms from an experiment that used a two-way ANOVA design for the deterministic stimulus inputs with inter-stimulus time intervals chosen from Poisson distributions of equal intensity.展开更多
Multi-source domain adaptation utilizes multiple source domains to learn the knowledge and transfers it to an unlabeled target domain.To address the problem,most of the existing methods aim to minimize the domain shif...Multi-source domain adaptation utilizes multiple source domains to learn the knowledge and transfers it to an unlabeled target domain.To address the problem,most of the existing methods aim to minimize the domain shift by auxiliary distribution alignment objectives,which reduces the effect of domain-specific features.However,without explicitly modeling the domain-specific features,it is not easy to guarantee that the domain-invariant representation extracted from input domains contains domain-specific information as few as possible.In this work,we present a different perspective on MSDA,which employs the idea of feature elimination to reduce the influence of domain-specific features.We design two different ways to extract domain-specific features and total features and construct the domain-invariant representations by eliminating the domain-specific features from total features.The experimental results on different domain adaptation datasets demonstrate the effectiveness of our method and the generalization ability of our model.展开更多
As a core component of power systems, the operational status of transformers directly affects grid stability. To address the problem of “domain shift” in cross-domain fault diagnosis, this paper proposes a memory-en...As a core component of power systems, the operational status of transformers directly affects grid stability. To address the problem of “domain shift” in cross-domain fault diagnosis, this paper proposes a memory-enhanced dual-stream network (MemFuse-DSN). The method reconstructs the feature space by selecting and enhancing multi-source domain samples based on similarity metrics. An adaptive weighted dual-stream architecture is designed, integrating gradient reversal and orthogonality constraints to achieve efficient feature alignment. In addition, a novel dual dynamic memory module is introduced: the task memory bank is used to store high-confidence class prototype information, and adopts an exponential moving average (EMA) strategy to ensure the smooth evolution of prototypes over time;the domain memory bank is periodically updated and clusters potential noisy features, dynamically tracking domain shift trends, thereby optimizing the decoupled feature learning process. Experimental validation was conducted on a ±110 kV transformer vibration testing platform using typical fault types including winding looseness, core looseness, and compound faults. The results show that the proposed method achieves a fault diagnosis accuracy of 99.2%, providing a highly generalizable solution for the intelligent operation and maintenance of power equipment.展开更多
The goal of decentralized multi-source domain adaptation is to conduct unsupervised multi-source domain adaptation in a data decentralization scenario. The challenge of data decentralization is that the source domains...The goal of decentralized multi-source domain adaptation is to conduct unsupervised multi-source domain adaptation in a data decentralization scenario. The challenge of data decentralization is that the source domains and target domain lack cross-domain collaboration during training. On the unlabeled target domain, the target model needs to transfer supervision knowledge with the collaboration of source models, while the domain gap will lead to limited adaptation performance from source models. On the labeled source domain, the source model tends to overfit its domain data in the data decentralization scenario, which leads to the negative transfer problem. For these challenges, we propose dual collaboration for decentralized multi-source domain adaptation by training and aggregating the local source models and local target model in collaboration with each other. On the target domain, we train the local target model by distilling supervision knowledge and fully using the unlabeled target domain data to alleviate the domain shift problem with the collaboration of local source models. On the source domain, we regularize the local source models in collaboration with the local target model to overcome the negative transfer problem. This forms a dual collaboration between the decentralized source domains and target domain, which improves the domain adaptation performance under the data decentralization scenario. Extensive experiments indicate that our method outperforms the state-of-the-art methods by a large margin on standard multi-source domain adaptation datasets.展开更多
Emotion recognition from physiological signals(ERPS)has drawn tremendous attention and can be potentially applied to numerous fields.Since physiological signals are nonstationary time series with high sampling frequen...Emotion recognition from physiological signals(ERPS)has drawn tremendous attention and can be potentially applied to numerous fields.Since physiological signals are nonstationary time series with high sampling frequency,it is challenging to directly extract features from them.Additionally,there are 2 major challenges in ERPS:(a)how to adequately capture the correlations between physiological signals at different times and between different types of physiological signals and(b)how to effectively minimize the negative effect caused by temporal covariate shift(TCS).To tackle these problems,we propose a domain generalization and residual network-based approach for emotion recognition from physiological signals(DGR-ERPS).We first pre-extract time-and frequency-domain features from the original time series to compose a new time series.Then,in order to fully extract the correlation information of different physiological signals,these time series are converted into 3D image data to serve as input for a residual-based feature encoder(RBFE).In addition,we introduce a domain generalization-based technique to mitigate the issue posed by TCS.We have conducted extensive experiments on 2 real-world datasets,and the results indicate that our DGR-ERPS achieves superior performance under both TCS and non-TCS scenarios.展开更多
We study a class of Dirichlet functions obtained as analytic continuation across the line of convergence of Dirichlet series which can be written as Euler products. This class includes that of Dirichlet L-functions. T...We study a class of Dirichlet functions obtained as analytic continuation across the line of convergence of Dirichlet series which can be written as Euler products. This class includes that of Dirichlet L-functions. The problem of the existence of multiple zeros for this last class is outstanding. It is tacitly accepted, yet not proved that the Riemann Zeta function, which belongs to this class, does not possess multiple zeros. In a previous study we provided an example of Dirichlet function having double zeros, but that function is not an Euler product function. In this paper we deal with Euler product functions and by using the geometric properties of the mapping realized by these functions, we tackle the problem of the multiplicity of their zeros.展开更多
We study the initial-boundary value problem of the Navier-Stokes equations for incompressible fluids in a general domain in R^n with compact and smooth boundary,subject to the kinematic and vorticity boundary conditio...We study the initial-boundary value problem of the Navier-Stokes equations for incompressible fluids in a general domain in R^n with compact and smooth boundary,subject to the kinematic and vorticity boundary conditions on the non-flat boundary.We observe that,under the nonhomogeneous boundary conditions,the pressure p can be still recovered by solving the Neumann problem for the Poisson equation.Then we establish the well-posedness of the unsteady Stokes equations and employ the solution to reduce our initial-boundary value problem into an initial-boundary value problem with absolute boundary conditions.Based on this,we first establish the well-posedness for an appropriate local linearized problem with the absolute boundary conditions and the initial condition(without the incompressibility condition),which establishes a velocity mapping.Then we develop apriori estimates for the velocity mapping,especially involving the Sobolev norm for the time-derivative of the mapping to deal with the complicated boundary conditions,which leads to the existence of the fixed point of the mapping and the existence of solutions to our initial-boundary value problem.Finally,we establish that,when the viscosity coefficient tends zero,the strong solutions of the initial-boundary value problem in R^n(n≥3)with nonhomogeneous vorticity boundary condition converge in L^2 to the corresponding Euler equations satisfying the kinematic condition.展开更多
Intelligent machinery fault diagnosis methods have been popularly and successfully developed in the past decades,and the vibration acceleration data collected by contact accelerometers have been widely investigated.In...Intelligent machinery fault diagnosis methods have been popularly and successfully developed in the past decades,and the vibration acceleration data collected by contact accelerometers have been widely investigated.In many industrial scenarios,contactless sensors are more preferred.The event camera is an emerging bio-inspired technology for vision sensing,which asynchronously records per-pixel brightness change polarity with high temporal resolution and low latency.It offers a promising tool for contactless machine vibration sensing and fault diagnosis.However,the dynamic vision-based methods suffer from variations of practical factors such as camera position,machine operating condition,etc.Furthermore,as a new sensing technology,the labeled dynamic vision data are limited,which generally cannot cover a wide range of machine fault modes.Aiming at these challenges,a novel dynamic vision-based machinery fault diagnosis method is proposed in this paper.It is motivated to explore the abundant vibration acceleration data for enhancing the dynamic vision-based model performance.A crossmodality feature alignment method is thus proposed with deep adversarial neural networks to achieve fault diagnosis knowledge transfer.An event erasing method is further proposed for improving model robustness against variations.The proposed method can effectively identify unseen fault mode with dynamic vision data.Experiments on two rotating machine monitoring datasets are carried out for validations,and the results suggest the proposed method is promising for generalized contactless machinery fault diagnosis.展开更多
The phenomenon of extinction is an important property of solutions for many evolutionary equa-tions. In this paper, a numerical simulation for computing the extinction time of nonnegative solu-tions for some nonlinear...The phenomenon of extinction is an important property of solutions for many evolutionary equa-tions. In this paper, a numerical simulation for computing the extinction time of nonnegative solu-tions for some nonlinear parabolic equations on general domains is presented. The solution algo-rithm utilizes the Donor-cell scheme in space and Euler’s method in time. Finally, we will give some numerical experiments to illustrate our algorithm.展开更多
基金supported by the National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(No.RS-2024-00406320)the Institute of Information&Communica-tions Technology Planning&Evaluation(IITP)-Innovative Human Resource Development for Local Intellectualization Program Grant funded by the Korea government(MSIT)(IITP-2026-RS-2023-00259678).
文摘Domain adaptation aims to reduce the distribution gap between the training data(source domain)and the target data.This enables effective predictions even for domains not seen during training.However,most conventional domain adaptation methods assume a single source domain,making them less suitable for modern deep learning settings that rely on diverse and large-scale datasets.To address this limitation,recent research has focused on Multi-Source Domain Adaptation(MSDA),which aims to learn effectively from multiple source domains.In this paper,we propose Efficient Domain Transition for Multi-source(EDTM),a novel and efficient framework designed to tackle two major challenges in existing MSDA approaches:(1)integrating knowledge across different source domains and(2)aligning label distributions between source and target domains.EDTM leverages an ensemble-based classifier expert mechanism to enhance the contribution of source domains that are more similar to the target domain.To further stabilize the learning process and improve performance,we incorporate imitation learning into the training of the target model.In addition,Maximum Classifier Discrepancy(MCD)is employed to align class-wise label distributions between the source and target domains.Experiments were conducted using Digits-Five,one of the most representative benchmark datasets for MSDA.The results show that EDTM consistently outperforms existing methods in terms of average classification accuracy.Notably,EDTM achieved significantly higher performance on target domains such as Modified National Institute of Standards and Technolog with blended background images(MNIST-M)and Street View House Numbers(SVHN)datasets,demonstrating enhanced generalization compared to baseline approaches.Furthermore,an ablation study analyzing the contribution of each loss component validated the effectiveness of the framework,highlighting the importance of each module in achieving optimal performance.
基金supported by the National Natural Science Foundation of China(62101575)the Research Project of NUDT(ZK22-57)the Self-directed Project of State Key Laboratory of High Performance Computing(202101-16).
文摘Automatically recognizing radar emitters from com-plex electromagnetic environments is important but non-trivial.Moreover,the changing electromagnetic environment results in inconsistent signal distribution in the real world,which makes the existing approaches perform poorly for recognition tasks in different scenes.In this paper,we propose a domain generaliza-tion framework is proposed to improve the adaptability of radar emitter signal recognition in changing environments.Specifically,we propose an end-to-end denoising based domain-invariant radar emitter recognition network(DDIRNet)consisting of a denoising model and a domain invariant representation learning model(IRLM),which mutually benefit from each other.For the signal denoising model,a loss function is proposed to match the feature of the radar signals and guarantee the effectiveness of the model.For the domain invariant representation learning model,contrastive learning is introduced to learn the cross-domain feature by aligning the source and unseen domain distri-bution.Moreover,we design a data augmentation method that improves the diversity of signal data for training.Extensive experiments on classification have shown that DDIRNet achieves up to 6.4%improvement compared with the state-of-the-art radar emitter recognition methods.The proposed method pro-vides a promising direction to solve the radar emitter signal recognition problem.
基金Supported by the National Natural Science Foundation of China(No.62001313)the Key Project of Liaoning Provincial Department of Science and Technology(No.2021JH2/10300134,2022JH1/10500004)。
文摘In the realm of medical image segmentation,particularly in cardiac magnetic resonance imaging(MRI),achieving robust performance with limited annotated data is a significant challenge.Performance often degrades when faced with testing scenarios from unknown domains.To address this problem,this paper proposes a novel semi-supervised approach for cardiac magnetic resonance image segmentation,aiming to enhance predictive capabilities and domain generalization(DG).This paper establishes an MT-like model utilizing pseudo-labeling and consistency regularization from semi-supervised learning,and integrates uncertainty estimation to improve the accuracy of pseudo-labels.Additionally,to tackle the challenge of domain generalization,a data manipulation strategy is introduced,extracting spatial and content-related information from images across different domains,enriching the dataset with a multi-domain perspective.This papers method is meticulously evaluated on the publicly available cardiac magnetic resonance imaging dataset M&Ms,validating its effectiveness.Comparative analyses against various methods highlight the out-standing performance of this papers approach,demonstrating its capability to segment cardiac magnetic resonance images in previously unseen domains even with limited annotated data.
基金supported by the National Natural Science Foundation of China(62322315,61873237)the Zhejiang Provincial Natural Science Foundation of China(LR22F030003)+1 种基金supported by Research Grant Council of Hong Kong(11201023,11202224)Hong Kong Innovation and Technology Commission(InnoHK Project CIMDA).
文摘In actual industrial scenarios,the variation of operating conditions,the existence of data noise,and failure of measurement equipment will inevitably affect the distribution of perceptive data.Deep learning-based fault diagnosis algorithms strongly rely on the assumption that source and target data are independent and identically distributed,and the learned diagnosis knowledge is difficult to generalize to out-of-distribution data.Domain generalization(DG)aims to achieve the generalization of arbitrary target domain data by using only limited source domain data for diagnosis model training.The research of DG for fault diagnosis has made remarkable progress in recent years and lots of achievements have been obtained.In this article,for the first time a comprehensive literature review on DG for fault diagnosis from a learning mechanism-oriented perspective is provided to summarize the development in recent years.Specifically,we first conduct a comprehensive review on existing methods based on the similarity of basic principles and design motivations.Then,the recent trend of DG for fault diagnosis is also analyzed.Finally,the existing problems and future prospect is performed.
文摘Single-source Domain Generalization(SDG)is a promising yet challenging technology that aims to transfer knowledge from a singular source domain to multiple and unfamiliar target domains.Existing SDG methods typically rely on domain expansion to implement data variation and broaden the coverage of the training domain.However,due to the lack of proper semantic consistency and sample diversity constraints,these methods have limited improvement in generalization performance for most practical applications.In this paper,we propose a Causality-Aware Single-source Domain Generalization(CASDG)method to utilize both semantic consistency and diversity during the data transformation process.First,a causality-aware module is designed to accurately measure the causal effect between latent features and labels.Then,we introduce a causal domain expansion module,which utilizes the causal effect matrix as a semantic consistency constraint and mutual information as a sample diversity constraint.These two constraints are jointly used to encourage the style transformer to generate new auxiliary samples that are undeviated from the original samples.The image classification model using our method can produce the best classification performance for unknown domain data compared to the state-of-the-art methods.
基金supported by the Deputyship for Research&Innovation,Ministry of Education in Saudi Arabia(Project No.MoE-IF-UJ-R2-22-04220773-1).
文摘Domain randomization is a widely adopted technique in deep reinforcement learning(DRL)to improve agent generalization by exposing policies to diverse environmental conditions.This paper investigates the impact of different reset strategies,normal,non-randomized,and randomized,on agent performance using the Deep Deterministic Policy Gradient(DDPG)and Twin Delayed DDPG(TD3)algorithms within the CarRacing-v2 environment.Two experimental setups were conducted:an extended training regime with DDPG for 1000 steps per episode across 1000 episodes,and a fast execution setup comparing DDPG and TD3 for 30 episodes with 50 steps per episode under constrained computational resources.A step-based reward scaling mechanism was applied under the randomized reset condition to promote broader state exploration.Experimental results showthat randomized resets significantly enhance learning efficiency and generalization,with DDPG demonstrating superior performance across all reset strategies.In particular,DDPG combined with randomized resets achieves the highest smoothed rewards(reaching approximately 15),best stability,and fastest convergence.These differences are statistically significant,as confirmed by t-tests:DDPG outperforms TD3 under randomized(t=−101.91,p<0.0001),normal(t=−21.59,p<0.0001),and non-randomized(t=−62.46,p<0.0001)reset conditions.The findings underscore the critical role of reset strategy and reward shaping in enhancing the robustness and adaptability of DRL agents in continuous control tasks,particularly in environments where computational efficiency and training stability are crucial.
基金supported by the Pre-research Project on Civil Aerospace Technologies of China National Space Administration(No.D010301).
文摘Learning domain-invariant feature representations is critical to alleviate the distribution differences between training and testing domains.The existing mainstream domain generalization approaches primarily pursue to align the across-domain distributions to extract the transferable feature representations.However,these representations may be insufficient and unstable.Moreover,these networks may also undergo catastrophic forgetting because the previous learned knowledge is replaced by the new learned knowledge.To cope with these issues,we propose a novel causality-based contrastive incremental learning model for domain generalization,which mainly includes three components:(1)intra-domain causal factorization,(2)inter-domain Mahalanobis similarity metric,and(3)contrastive knowledge distillation.The model extracts intra and inter domain-invariant knowledge to improve model generalization.Specifically,we first introduce a causal factorization to extract intra-domain invariant knowledge.Then,we design a Mahalanobis similarity metric to extract common inter-domain invariant knowledge.Finally,we propose a contrastive knowledge distillation with exponential moving average to distill model parameters in a smooth way to preserve the previous learned knowledge and mitigate model forgetting.Extensive experiments on several domain generalization benchmarks prove that our model achieves the state-of-the-art results,which sufficiently show the effectiveness of our model.
文摘We dealt in a series of previous publications with some geometric aspects of the mappings by functions obtained as analytic continuations to the whole complex plane of general Dirichlet series. Pictures illustrating those aspects contain a lot of other information which has been waiting for a rigorous proof. Such a task is partially fulfilled in this paper, where we succeeded among other things, to prove a theorem about general Dirichlet series having as corollary the Speiser’s theorem. We have also proved that those functions do not possess multiple zeros of order higher than 2 and the double zeros have very particular locations. Moreover, their derivatives have only simple zeros. With these results at hand, we revisited GRH for a simplified proof.
基金Supported by the Scientific and Technological Innovation 2030—Major Project of"New Generation Artificial Intelligence"(2020AAA0109300)。
文摘Smart manufacturing suffers from the heterogeneity of local data distribution across parties,mutual information silos and lack of privacy protection in the process of industry chain collaboration.To address these problems,we propose a federated domain adaptation algorithm based on knowledge distillation and contrastive learning.Knowledge distillation is used to extract transferable integration knowledge from the different source domains and the quality of the extracted integration knowledge is used to assign reasonable weights to each source domain.A more rational weighted average aggregation is used in the aggregation phase of the center server to optimize the global model,while the local model of the source domain is trained with the help of contrastive learning to constrain the local model optimum towards the global model optimum,mitigating the inherent heterogeneity between local data.Our experiments are conducted on the largest domain adaptation dataset,and the results show that compared with other traditional federated domain adaptation algorithms,the algorithm we proposed trains a more accurate model,requires fewer communication rounds,makes more effective use of imbalanced data in the industrial area,and protects data privacy.
基金The authors gratefully acknowledge the financial support from the National Natural Science Foundation of China under Grant 81871395.
文摘Previous studies have already shown that Raman spectroscopy can be used in the encoding of suspension array technology.However,almost all existing convolutional neural network-based decoding approaches rely on supervision with ground truth,and may not be well generalized to unseen datasets,which were collected under different experimental conditions,applying with the same coded material.In this study,we propose an improved model based on CyCADA,named as Detail constraint Cycle Domain Adaptive Model(DCDA).DCDA implements the clasification of unseen datasets through domain adaptation,adapts representations at the encode level with decoder-share,and enforces coding features while leveraging a feat loss.To improve detailed structural constraints,DCDA takes downsample connection and skips connection.Our model improves the poor generalization of existing models and saves the cost of the labeling process for unseen target datasets.Compared with other models,extensive experiments and ablation studies show the superiority of DCDA in terms of classification stability and generalization.The model proposed by the research achieves a classification with an accuracy of 100%when applied in datasets,in which the spectrum in the source domain is far less than the target domain.
文摘In this paper,we first obtain a unified integral representation on the analytic varieties of the general bounded domain in Stein manifolds(the two types bounded domains in[3]are regarded as its special cases).Secondly we get the integral formulas of the solution of∂-equation.And we use a new and unique method to give a uniform estimate of the solution of∂-equation,which is different from Henkin's method.
文摘Analysis of functional MRI (fMRI) blood oxygenation level dependent (BOLD) data is typically carried out in the time domain where the data has a high temporal correlation. These analyses usually employ parametric models of the hemodynamic response function (HRF) where either pre-whitening of the data is attempted or autoregressive (AR) models are employed to model the noise. Statistical analysis then proceeds via regression of the convolution of the HRF with the input stimuli. This approach has limitations when considering that the time series collected are embedded in a brain image in which the AR model order may vary and pre-whitening techniques may be insufficient for handling faster sampling times. However fMRI data can be analyzed in the Fourier domain where the assumptions made as to the structure of the noise can be less restrictive and hypothesis tests are straightforward for single subject analysis, especially useful in a clinical setting. This allows for experiments that can have both fast temporal sampling and event-related designs where stimuli can be closely spaced in time. Equally important, statistical analysis in the Fourier domain focuses on hypothesis tests based on nonparametric estimates of the hemodynamic transfer function (HRF in the frequency domain). This is especially important for experimental designs involving multiple states (drug or stimulus induced) that may alter the form of the response function. In this context a univariate general linear model in the Fourier domain has been applied to analyze BOLD data sampled at a rate of 400 ms from an experiment that used a two-way ANOVA design for the deterministic stimulus inputs with inter-stimulus time intervals chosen from Poisson distributions of equal intensity.
基金supported by the National Natural Science Foundation of China(NSFC)(Grant Nos.61876130 and 61932009).
文摘Multi-source domain adaptation utilizes multiple source domains to learn the knowledge and transfers it to an unlabeled target domain.To address the problem,most of the existing methods aim to minimize the domain shift by auxiliary distribution alignment objectives,which reduces the effect of domain-specific features.However,without explicitly modeling the domain-specific features,it is not easy to guarantee that the domain-invariant representation extracted from input domains contains domain-specific information as few as possible.In this work,we present a different perspective on MSDA,which employs the idea of feature elimination to reduce the influence of domain-specific features.We design two different ways to extract domain-specific features and total features and construct the domain-invariant representations by eliminating the domain-specific features from total features.The experimental results on different domain adaptation datasets demonstrate the effectiveness of our method and the generalization ability of our model.
基金supported by the State Grid Shandong Electric Power Company Project(Grant Number SGSDJX00BDJS2400388).
文摘As a core component of power systems, the operational status of transformers directly affects grid stability. To address the problem of “domain shift” in cross-domain fault diagnosis, this paper proposes a memory-enhanced dual-stream network (MemFuse-DSN). The method reconstructs the feature space by selecting and enhancing multi-source domain samples based on similarity metrics. An adaptive weighted dual-stream architecture is designed, integrating gradient reversal and orthogonality constraints to achieve efficient feature alignment. In addition, a novel dual dynamic memory module is introduced: the task memory bank is used to store high-confidence class prototype information, and adopts an exponential moving average (EMA) strategy to ensure the smooth evolution of prototypes over time;the domain memory bank is periodically updated and clusters potential noisy features, dynamically tracking domain shift trends, thereby optimizing the decoupled feature learning process. Experimental validation was conducted on a ±110 kV transformer vibration testing platform using typical fault types including winding looseness, core looseness, and compound faults. The results show that the proposed method achieves a fault diagnosis accuracy of 99.2%, providing a highly generalizable solution for the intelligent operation and maintenance of power equipment.
基金Project supported by the National Nature Science Foundation of China (Nos. 61876130 and 61932009)the Starry Night Science Fund of Zhejiang University Shanghai Institute for Advanced Study。
文摘The goal of decentralized multi-source domain adaptation is to conduct unsupervised multi-source domain adaptation in a data decentralization scenario. The challenge of data decentralization is that the source domains and target domain lack cross-domain collaboration during training. On the unlabeled target domain, the target model needs to transfer supervision knowledge with the collaboration of source models, while the domain gap will lead to limited adaptation performance from source models. On the labeled source domain, the source model tends to overfit its domain data in the data decentralization scenario, which leads to the negative transfer problem. For these challenges, we propose dual collaboration for decentralized multi-source domain adaptation by training and aggregating the local source models and local target model in collaboration with each other. On the target domain, we train the local target model by distilling supervision knowledge and fully using the unlabeled target domain data to alleviate the domain shift problem with the collaboration of local source models. On the source domain, we regularize the local source models in collaboration with the local target model to overcome the negative transfer problem. This forms a dual collaboration between the decentralized source domains and target domain, which improves the domain adaptation performance under the data decentralization scenario. Extensive experiments indicate that our method outperforms the state-of-the-art methods by a large margin on standard multi-source domain adaptation datasets.
基金supported in part by the National Natural Science Foundation of China under grants 62236005,61936004,and U1913602.
文摘Emotion recognition from physiological signals(ERPS)has drawn tremendous attention and can be potentially applied to numerous fields.Since physiological signals are nonstationary time series with high sampling frequency,it is challenging to directly extract features from them.Additionally,there are 2 major challenges in ERPS:(a)how to adequately capture the correlations between physiological signals at different times and between different types of physiological signals and(b)how to effectively minimize the negative effect caused by temporal covariate shift(TCS).To tackle these problems,we propose a domain generalization and residual network-based approach for emotion recognition from physiological signals(DGR-ERPS).We first pre-extract time-and frequency-domain features from the original time series to compose a new time series.Then,in order to fully extract the correlation information of different physiological signals,these time series are converted into 3D image data to serve as input for a residual-based feature encoder(RBFE).In addition,we introduce a domain generalization-based technique to mitigate the issue posed by TCS.We have conducted extensive experiments on 2 real-world datasets,and the results indicate that our DGR-ERPS achieves superior performance under both TCS and non-TCS scenarios.
文摘We study a class of Dirichlet functions obtained as analytic continuation across the line of convergence of Dirichlet series which can be written as Euler products. This class includes that of Dirichlet L-functions. The problem of the existence of multiple zeros for this last class is outstanding. It is tacitly accepted, yet not proved that the Riemann Zeta function, which belongs to this class, does not possess multiple zeros. In a previous study we provided an example of Dirichlet function having double zeros, but that function is not an Euler product function. In this paper we deal with Euler product functions and by using the geometric properties of the mapping realized by these functions, we tackle the problem of the multiplicity of their zeros.
基金supported in part by the National Science Foundation under Grants DMS-0807551,DMS-0720925,and DMS-0505473the Natural Science Foundationof China(10728101)supported in part by EPSRC grant EP/F029578/1
文摘We study the initial-boundary value problem of the Navier-Stokes equations for incompressible fluids in a general domain in R^n with compact and smooth boundary,subject to the kinematic and vorticity boundary conditions on the non-flat boundary.We observe that,under the nonhomogeneous boundary conditions,the pressure p can be still recovered by solving the Neumann problem for the Poisson equation.Then we establish the well-posedness of the unsteady Stokes equations and employ the solution to reduce our initial-boundary value problem into an initial-boundary value problem with absolute boundary conditions.Based on this,we first establish the well-posedness for an appropriate local linearized problem with the absolute boundary conditions and the initial condition(without the incompressibility condition),which establishes a velocity mapping.Then we develop apriori estimates for the velocity mapping,especially involving the Sobolev norm for the time-derivative of the mapping to deal with the complicated boundary conditions,which leads to the existence of the fixed point of the mapping and the existence of solutions to our initial-boundary value problem.Finally,we establish that,when the viscosity coefficient tends zero,the strong solutions of the initial-boundary value problem in R^n(n≥3)with nonhomogeneous vorticity boundary condition converge in L^2 to the corresponding Euler equations satisfying the kinematic condition.
基金supported by the National Science Fund for Distinguished Young Scholars of China(52025056)the China Postdoctoral Science Foundation(2023M732789)+1 种基金the China Postdoctoral Innovative Talents Support Program(BX20230290)the Fundamental Research Funds for the Central Universities(xzy012022062).
文摘Intelligent machinery fault diagnosis methods have been popularly and successfully developed in the past decades,and the vibration acceleration data collected by contact accelerometers have been widely investigated.In many industrial scenarios,contactless sensors are more preferred.The event camera is an emerging bio-inspired technology for vision sensing,which asynchronously records per-pixel brightness change polarity with high temporal resolution and low latency.It offers a promising tool for contactless machine vibration sensing and fault diagnosis.However,the dynamic vision-based methods suffer from variations of practical factors such as camera position,machine operating condition,etc.Furthermore,as a new sensing technology,the labeled dynamic vision data are limited,which generally cannot cover a wide range of machine fault modes.Aiming at these challenges,a novel dynamic vision-based machinery fault diagnosis method is proposed in this paper.It is motivated to explore the abundant vibration acceleration data for enhancing the dynamic vision-based model performance.A crossmodality feature alignment method is thus proposed with deep adversarial neural networks to achieve fault diagnosis knowledge transfer.An event erasing method is further proposed for improving model robustness against variations.The proposed method can effectively identify unseen fault mode with dynamic vision data.Experiments on two rotating machine monitoring datasets are carried out for validations,and the results suggest the proposed method is promising for generalized contactless machinery fault diagnosis.
文摘The phenomenon of extinction is an important property of solutions for many evolutionary equa-tions. In this paper, a numerical simulation for computing the extinction time of nonnegative solu-tions for some nonlinear parabolic equations on general domains is presented. The solution algo-rithm utilizes the Donor-cell scheme in space and Euler’s method in time. Finally, we will give some numerical experiments to illustrate our algorithm.
