Climate downscaling is used to transform large-scale meteorological data into small-scale data with enhanced detail,which finds wide applications in climate modeling,numerical weather forecasting,and renewable energy....Climate downscaling is used to transform large-scale meteorological data into small-scale data with enhanced detail,which finds wide applications in climate modeling,numerical weather forecasting,and renewable energy.Although deeplearning-based downscaling methods effectively capture the complex nonlinear mapping between meteorological data of varying scales,the supervised deep-learning-based downscaling methods suffer from insufficient high-resolution data in practice,and unsupervised methods struggle with accurately inferring small-scale specifics from limited large-scale inputs due to small-scale uncertainty.This article presents DualDS,a dual-learning framework utilizing a Generative Adversarial Network–based neural network and subgrid-scale auxiliary information for climate downscaling.Such a learning method is unified in a two-stream framework through up-and downsamplers,where the downsampler is used to simulate the information loss process during the upscaling,and the upsampler is used to reconstruct lost details and correct errors incurred during the upscaling.This dual learning strategy can eliminate the dependence on high-resolution ground truth data in the training process and refine the downscaling results by constraining the mapping process.Experimental findings demonstrate that DualDS is comparable to several state-of-the-art deep learning downscaling approaches,both qualitatively and quantitatively.Specifically,for a single surface-temperature data downscaling task,our method is comparable with other unsupervised algorithms with the same dataset,and we can achieve a 0.469 dB higher peak signal-to-noise ratio,0.017 higher structural similarity,0.08 lower RMSE,and the best correlation coefficient.In summary,this paper presents a novel approach to addressing small-scale uncertainty issues in unsupervised downscaling processes.展开更多
The unsupervised vehicle re-identification task aims at identifying specific vehicles in surveillance videos without utilizing annotation information.Due to the higher similarity in appearance between vehicles compare...The unsupervised vehicle re-identification task aims at identifying specific vehicles in surveillance videos without utilizing annotation information.Due to the higher similarity in appearance between vehicles compared to pedestrians,pseudo-labels generated through clustering are ineffective in mitigating the impact of noise,and the feature distance between inter-class and intra-class has not been adequately improved.To address the aforementioned issues,we design a dual contrastive learning method based on knowledge distillation.During each iteration,we utilize a teacher model to randomly partition the entire dataset into two sub-domains based on clustering pseudo-label categories.By conducting contrastive learning between the two student models,we extract more discernible vehicle identity cues to improve the problem of imbalanced data distribution.Subsequently,we propose a context-aware pseudo label refinement strategy that leverages contextual features by progressively associating granularity information from different bottleneck blocks.To produce more trustworthy pseudo-labels and lessen noise interference during the clustering process,the context-aware scores are obtained by calculating the similarity between global features and contextual ones,which are subsequently added to the pseudo-label encoding process.The proposed method has achieved excellent performance in overcoming label noise and optimizing data distribution through extensive experimental results on publicly available datasets.展开更多
Sum-product networks(SPNs)are an expressive deep probabilistic architecture with solid theoretical foundations,which allows tractable and exact inference.SPNs always act as black-box inference machine in many artifici...Sum-product networks(SPNs)are an expressive deep probabilistic architecture with solid theoretical foundations,which allows tractable and exact inference.SPNs always act as black-box inference machine in many artificial intelligence tasks.Due to their recursive definition,SPNs can also be naturally employed as hierarchical feature extractors.Recently,SPNs have been successfully employed as autoencoder framework in representation learning.However,SPNs autoencoder ignores the model structural duality and trains the models separately and independently.In this work,we propose a Dual-SPNs autoencoder which designs two SPNs autoencoders to compose as a dual form.This approach trains the models simultaneously,and explicitly exploits the structural duality between them to enhance the training process.Experimental results on several multilabel classification problems demonstrate that Dual-SPNs autoencoder is very competitive against with state-of-the-art autoencoder architectures.展开更多
Control of the fed-batch ethanol fermentation processes to produce maximum product ethanol is one of the key issues in the bioreactor system.However,ethanol fermentation processes exhibit complex behavior and nonlinea...Control of the fed-batch ethanol fermentation processes to produce maximum product ethanol is one of the key issues in the bioreactor system.However,ethanol fermentation processes exhibit complex behavior and nonlinear dynamics with respect to the cell mass,substrate,feed-rate,etc.An improved dual heuristic programming algorithm based on the least squares temporal difference with gradient correction(LSTDC) algorithm(LSTDC-DHP) is proposed to solve the learning control problem of a fed-batch ethanol fermentation process.As a new algorithm of adaptive critic designs,LSTDC-DHP is used to realize online learning control of chemical dynamical plants,where LSTDC is commonly employed to approximate the value functions.Application of the LSTDC-DHP algorithm to ethanol fermentation process can realize efficient online learning control in continuous spaces.Simulation results demonstrate the effectiveness of LSTDC-DHP,and show that LSTDC-DHP can obtain the near-optimal feed rate trajectory faster than other-based algorithms.展开更多
Financial technology(FinTech)has emerged as a key policy instrument for transforming economic development models.Whether it can energize the low-carbon economy and enhance carbon emission efficiency(CEE)has drawn incr...Financial technology(FinTech)has emerged as a key policy instrument for transforming economic development models.Whether it can energize the low-carbon economy and enhance carbon emission efficiency(CEE)has drawn increasing scholarly attention.Using panel data from 278 Chinese cities(2006–2021),this study constructs a quasi-natural experiment and applies a difference-in-differences(DID)model to evaluate the policy effects of FinTech on CEE.Mediation and moderation models further explore the mechanisms underlying this relationship.The findings reveal that FinTech significantly improves CEE,generating policy dividends that advance low-carbon development.This conclusion remains robust under dual machine learning causal inference,propensity score matching DID,and other robustness tests.Financing constraints,innovation level,and industrial structure upgrading fully mediate the effect of FinTech on CEE,where financing constraints and innovation show positive mediation effects,while industrial structure upgrading has a negative one.Moreover,financial agglomeration weakens FinTech’s positive influence on CEE,and the effects differ across regions,development levels,and resource endowments.This study broadens the analytical framework connecting FinTech and CEE and deepens understanding of the mechanisms linking financial innovation to lowcarbon transition.展开更多
Graphene grids exhibit exceptional loading capacity for macromolecules,single atoms,and nanoparticles,offering significant potential for exploring the structure and properties of various materials at the nanoscale.How...Graphene grids exhibit exceptional loading capacity for macromolecules,single atoms,and nanoparticles,offering significant potential for exploring the structure and properties of various materials at the nanoscale.However,challenges such as carbon film rupture,contamination,and uneven graphene film coverage frequently occur during grid fabrication.Here wepropose a dual-stage deep learning model integrating U-Net and an enhanced YOLO11 architecture,enabling efficient and accurate defect detection and graphene coverage quantification.A tailored data augmentation strategy expanded the initial defect dataset by more than an order of magnitude,which directly contributed to an overall 11.72%improvement across the model’s performance metrics.With the integration of the multi-scale convolutional attention(MSCA)module and the slicing-aided hyper inference(SAHI)method,the model achieved a 0.67%mean absolute percentage error(MAPE),while reducing the average detection time from 26.6 to 0.1 min per image.The proposed model holds strong potential for extension to various material characterization image analysis tasks,providing a scalable strategy for high-throughput image processing that bridges fundamental research with industrialscale applications.展开更多
基金supported by the following funding bodies:the National Key Research and Development Program of China(Grant No.2020YFA0608000)National Science Foundation of China(Grant Nos.42075142,42375148,42125503+2 种基金42130608)FY-APP-2022.0609,Sichuan Province Key Tech nology Research and Development project(Grant Nos.2024ZHCG0168,2024ZHCG0176,2023YFG0305,2023YFG-0124,and 23ZDYF0091)the CUIT Science and Technology Innovation Capacity Enhancement Program project(Grant No.KYQN202305)。
文摘Climate downscaling is used to transform large-scale meteorological data into small-scale data with enhanced detail,which finds wide applications in climate modeling,numerical weather forecasting,and renewable energy.Although deeplearning-based downscaling methods effectively capture the complex nonlinear mapping between meteorological data of varying scales,the supervised deep-learning-based downscaling methods suffer from insufficient high-resolution data in practice,and unsupervised methods struggle with accurately inferring small-scale specifics from limited large-scale inputs due to small-scale uncertainty.This article presents DualDS,a dual-learning framework utilizing a Generative Adversarial Network–based neural network and subgrid-scale auxiliary information for climate downscaling.Such a learning method is unified in a two-stream framework through up-and downsamplers,where the downsampler is used to simulate the information loss process during the upscaling,and the upsampler is used to reconstruct lost details and correct errors incurred during the upscaling.This dual learning strategy can eliminate the dependence on high-resolution ground truth data in the training process and refine the downscaling results by constraining the mapping process.Experimental findings demonstrate that DualDS is comparable to several state-of-the-art deep learning downscaling approaches,both qualitatively and quantitatively.Specifically,for a single surface-temperature data downscaling task,our method is comparable with other unsupervised algorithms with the same dataset,and we can achieve a 0.469 dB higher peak signal-to-noise ratio,0.017 higher structural similarity,0.08 lower RMSE,and the best correlation coefficient.In summary,this paper presents a novel approach to addressing small-scale uncertainty issues in unsupervised downscaling processes.
基金supported by the National Natural Science Foundation of China under Grant Nos.62461037,62076117 and 62166026the Jiangxi Provincial Natural Science Foundation under Grant Nos.20224BAB212011,20232BAB202051,20232BAB212008 and 20242BAB25078the Jiangxi Provincial Key Laboratory of Virtual Reality under Grant No.2024SSY03151.
文摘The unsupervised vehicle re-identification task aims at identifying specific vehicles in surveillance videos without utilizing annotation information.Due to the higher similarity in appearance between vehicles compared to pedestrians,pseudo-labels generated through clustering are ineffective in mitigating the impact of noise,and the feature distance between inter-class and intra-class has not been adequately improved.To address the aforementioned issues,we design a dual contrastive learning method based on knowledge distillation.During each iteration,we utilize a teacher model to randomly partition the entire dataset into two sub-domains based on clustering pseudo-label categories.By conducting contrastive learning between the two student models,we extract more discernible vehicle identity cues to improve the problem of imbalanced data distribution.Subsequently,we propose a context-aware pseudo label refinement strategy that leverages contextual features by progressively associating granularity information from different bottleneck blocks.To produce more trustworthy pseudo-labels and lessen noise interference during the clustering process,the context-aware scores are obtained by calculating the similarity between global features and contextual ones,which are subsequently added to the pseudo-label encoding process.The proposed method has achieved excellent performance in overcoming label noise and optimizing data distribution through extensive experimental results on publicly available datasets.
基金the National Natural Science Foundation of China(No.61472161)the Science&Technology Development Project of Jilin Province(Nos.20180101334JC and 20160520099JH)。
文摘Sum-product networks(SPNs)are an expressive deep probabilistic architecture with solid theoretical foundations,which allows tractable and exact inference.SPNs always act as black-box inference machine in many artificial intelligence tasks.Due to their recursive definition,SPNs can also be naturally employed as hierarchical feature extractors.Recently,SPNs have been successfully employed as autoencoder framework in representation learning.However,SPNs autoencoder ignores the model structural duality and trains the models separately and independently.In this work,we propose a Dual-SPNs autoencoder which designs two SPNs autoencoders to compose as a dual form.This approach trains the models simultaneously,and explicitly exploits the structural duality between them to enhance the training process.Experimental results on several multilabel classification problems demonstrate that Dual-SPNs autoencoder is very competitive against with state-of-the-art autoencoder architectures.
基金Supported by the National Natural Science Foundation of China(61573052)
文摘Control of the fed-batch ethanol fermentation processes to produce maximum product ethanol is one of the key issues in the bioreactor system.However,ethanol fermentation processes exhibit complex behavior and nonlinear dynamics with respect to the cell mass,substrate,feed-rate,etc.An improved dual heuristic programming algorithm based on the least squares temporal difference with gradient correction(LSTDC) algorithm(LSTDC-DHP) is proposed to solve the learning control problem of a fed-batch ethanol fermentation process.As a new algorithm of adaptive critic designs,LSTDC-DHP is used to realize online learning control of chemical dynamical plants,where LSTDC is commonly employed to approximate the value functions.Application of the LSTDC-DHP algorithm to ethanol fermentation process can realize efficient online learning control in continuous spaces.Simulation results demonstrate the effectiveness of LSTDC-DHP,and show that LSTDC-DHP can obtain the near-optimal feed rate trajectory faster than other-based algorithms.
基金supported by the National Natural Science Foundation of China[Grant No.72371179]the Social Science Foundation of Jiangsu Province[Grant Nos.25GLB009 and 25ZXZD010]+3 种基金the Talent Cultivation Foundation of Southwest University of Science and Technology[Grant No.25sx7003]the Key Laboratory of Green Materials Development and Innovation[Grant No.LSCL-2502]the Major Special Project of the Sichuan Provincial Philosophy and Social Science Foundation[Grant No.SCJJ24ZD30]the Collaborative Innovation Center for New Urbanization and Social Governance of Soochow University.
文摘Financial technology(FinTech)has emerged as a key policy instrument for transforming economic development models.Whether it can energize the low-carbon economy and enhance carbon emission efficiency(CEE)has drawn increasing scholarly attention.Using panel data from 278 Chinese cities(2006–2021),this study constructs a quasi-natural experiment and applies a difference-in-differences(DID)model to evaluate the policy effects of FinTech on CEE.Mediation and moderation models further explore the mechanisms underlying this relationship.The findings reveal that FinTech significantly improves CEE,generating policy dividends that advance low-carbon development.This conclusion remains robust under dual machine learning causal inference,propensity score matching DID,and other robustness tests.Financing constraints,innovation level,and industrial structure upgrading fully mediate the effect of FinTech on CEE,where financing constraints and innovation show positive mediation effects,while industrial structure upgrading has a negative one.Moreover,financial agglomeration weakens FinTech’s positive influence on CEE,and the effects differ across regions,development levels,and resource endowments.This study broadens the analytical framework connecting FinTech and CEE and deepens understanding of the mechanisms linking financial innovation to lowcarbon transition.
基金supported by National Key Research and Development Program of China(2024YFB4709300)the National Natural Science Foundation of China(No.52130501,52505289)+1 种基金Zhejiang provincial teams of leading talents in Innovation and Entrepreneurship(2024R01002)Guizhou Provincial Science and Technology Projects(XKBF[2025]014,BQW[2024]010).
文摘Graphene grids exhibit exceptional loading capacity for macromolecules,single atoms,and nanoparticles,offering significant potential for exploring the structure and properties of various materials at the nanoscale.However,challenges such as carbon film rupture,contamination,and uneven graphene film coverage frequently occur during grid fabrication.Here wepropose a dual-stage deep learning model integrating U-Net and an enhanced YOLO11 architecture,enabling efficient and accurate defect detection and graphene coverage quantification.A tailored data augmentation strategy expanded the initial defect dataset by more than an order of magnitude,which directly contributed to an overall 11.72%improvement across the model’s performance metrics.With the integration of the multi-scale convolutional attention(MSCA)module and the slicing-aided hyper inference(SAHI)method,the model achieved a 0.67%mean absolute percentage error(MAPE),while reducing the average detection time from 26.6 to 0.1 min per image.The proposed model holds strong potential for extension to various material characterization image analysis tasks,providing a scalable strategy for high-throughput image processing that bridges fundamental research with industrialscale applications.
