检索结果-内蒙古大学图书馆

27th International Conference on Pattern Recognition, ICPR 2024

作者： Li, Bowen Zhang, Jian Li, Jie Wu, Chentao Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai China Yancheng Blockchain Research Institute Yancheng China Blockchain Advanced Research Center Shanghai China Shanghai Key Laboratory of Trusted Data Circulation and Governance and Web3 Shanghai China

ISBN: (纸本)9783031781681

In response to legislation, companies are now mandated to honor the right to be forgotten by erasing user data. Consequently, it has become imperative to enable data removal in Vertical Federated Learning (VFL), where multiple parties provide private features for model training. In VFL, data removal, i.e., machine unlearning, requires removing specific features across all samples, ensuring privacy in federated learning. To address this challenge, we propose SecureCut, a novel Gradient Boosting Decision Tree (GBDT) framework that effectively enables both instance unlearning and feature unlearning without the need for retraining from scratch. Leveraging a robust GBDT structure, we enable effective data deletion while reducing degradation of model performance. Extensive experimental results on popular datasets demonstrate that our method achieves superior model utility and forgetfulness compared to the state-of-the-art methods. To the best of our knowledge, this is the first work that investigates machine unlearning in VFL scenarios. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Research and Implementation of OFC-Based Multicarrier Sliceable Bandwidth Variable Transponders in Hybrid Laser/Microwave Satellite Networks

引用

IEEE Access 2025年 13卷 93046-93056页

作者： Bin Wu Hao Zhang Qingchun Zhao Hebei Key Laboratory of Marine Perception Network and Data Processing School of Computer and Communication Engineering Northeastern University at Qinhuangdao Qinhuangdao China

In this paper, a multicarrier sliceable bandwidth variable transponder based on optical frequency comb (OFC) is proposed. The structure of the transponder’s internal switching unit is designed, and the principle of generating ultra-flat OFC based on a standalone dual-parallel Mach-Zehnder modulator (DPMZM) is analyzed. A scheme for dynamic allocation of multi-channel frequency slots and flexible resource configuration in hybrid laser/satellite networks is provided, and an experimental system is built to test the data transmission and bandwidth allocation capabilities of laser and microwave links. Experimental results indicate that the muticarrier generation unit can produce a 7-line flat OFC with a flatness of 1.6 dB, and the switching unit can achieve channel switching and frequency slot allocation, with a minimum slice width of 12.5 GHz. After passing through this node, the phase noise of the 2 GHz IF signal with frequency offsets at 10 kHz and 100 kHz are −59.28 dBc/Hz and −87.95 dBc/Hz, respectively, and the bit error rates (BERs) of the recovered baseband data from the laser and microwave links are less than $10^{-9}$ . These results validate the feasibility of the designed transponder to be applied in the key relay nodes of satellite networks. The proposed scheme can solve the problems of low spectrum utilization and the inability to dynamically allocate resources in traditional WDM satellite networks.

关键词： Satellites Transponders Masers Optical switches Frequency modulation Bandwidth Resource management Optical filters Microwave communication Optical fiber communication

来源：评论

学校读者我要写书评

暂无评论

Electroless plating method prepared hierarchical structure of nickel black membrane for solar-driven interfacial evaporation

引用

Chemical engineering Research and Design 2025年 218卷 587-596页

作者： He, Shuang Ma, keyi Zhang, Hao Zhang, Guoliang Yang, Longchao Zhou, Jinhao Wu, Wenxuan Sun, De Li, Bingbing Yue, Dongmin School of Chemical Engineering Changchun University of Technology 2055 Yanan Street Changchun130012 China Key Laboratory of Advanced Functional Polymer Membrane Materials of Jilin Province 2055 Yanan Street Changchun130012 China State Grid Jilin Electric Power Research Institute 4433 Renmin Street Changchun130021 China Advanced Institute of Materials Science Changchun University of Technology 2055 Yanan Street Changchun130012 China

Solar-driven interfacial evaporation technology has attracted significant attention because of its ability to concentrate thermal energy on the evaporation surface to minimize the heat loss. As a key component in the process of interfacial evaporation, the structure of the photothermal conversion membrane is one of the important factors influencing interfacial evaporation performance. Herein, a hierarchical nickel black membrane (H-Ni) was prepared by electroless plating for solar-driven interfacial evaporation. A layer of triangular nickel black particles was grown on the surface of the Ni foam skeleton, which enhances the absorption (93.87 %) through the light-trapping effect. The evaporation rate of H-Ni membrane achieved to 1.27 kg m−2 h−1 with a solar-vapor conversion efficiency of 80.22 % under 1 kW m−2 solar irradiation. This work provides a novel design of a hierarchical structure via a convenient way, which is conductive to promoting the widespread application of solar water treatment technology. © 2025 Institution of Chemical Engineers

关键词： Thermal evaporation

来源：评论

学校读者我要写书评

暂无评论

Forecasting of exchange rate time series based on event-aware transformer mode

引用

Soft Computing 2025年第6期29卷 3035-3045页

作者： Zhang, Siyi Che, Tong Zhu, Zhiliang Luo, Guoli Feng, Ping Virtual Reality and Interactive Techniques Institute East China Jiaotong University Nanchang330013 China Department of Engineering Mathematics University of Bristol Northern Ireland BristolBS8 1TH United Kingdom Guizhou Key Laboratory of Big Data Statistical Analysis Guizhou University of Finance and Economics Guiyang550025 China

Accurately forecasting exchange rate time series is crucial for effective risk management. The longer the forecast time window and the higher the frequency, the more valuable it is for managers to make timely decisions. In light of the task characteristics of exchange rate forecasting, this paper proposes an improved transformer-based model for addressing long sequence time series forecasting problems. Specifically, we introduce a position embedding method for events, which integrates the traditional coding context vector to perform joint location representation learning. Additionally, we explore a periodic enhancement method of features to jointly enhance the transformer-like model architecture’s ability to adaptively learn the long-term and hidden correlations of long sequence time series data. Extensive experimental results on the prediction of four exchange rate data for pairs of opening, high, low, and closing prices at a frequency of 5 min show that our model outperforms the prediction of the baseline method. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Risk management

来源：评论

学校读者我要写书评

暂无评论

Lightweight Remote Sensing Target Detection Algorithm Based on Improved YOLOv5 16th

Lightweight Remote Sensing Target Detection Algorithm Based ...

引用

16th International Conference on Genetic and Evolutionary Computing, ICGEC 2024

作者： Wang, Chaomin Dong, Fang Liang, Sheng Yu, Wanxin Wu, Kaixian Yu, Long Wenfeng, Wang School of Information Engineering Nanchang Institute of Technology Nanchang330099 China Jiangxi Province Key Laboratory of Smart Water Conservancy Nanchang Institute of Technology Nanchang330099 China School of Statistics and Data Science Jiangxi University of Finance and Economics Nanchang330013 China

ISBN: (纸本)9789819615346

Aiming at the problems of large memory space occupied by remote sensing image detection models in complex scenes, high computational complexity, and detection speed that is difficult to meet the real-time target detection requirements, a model compression algorithm based on dynamic detection head and channel pruning is proposed. Based on YOLOv5s, the head of the original YOLOv5 is replaced by the DyHead detection head, which improves the detection performance of the model by combining and unifying the three attention mechanisms of scale, space and task. Secondly, the Focaler-IoU loss function is introduced to further improve the predictive regression ability of the model. Finally, the channel pruning algorithm is used to eliminate the redundant parameters in the network to further compress the parameters and computation of the model, and the accuracy of the pruned model is improved by model fine-tuning. The experimental results show that there is a large improvement on the DOTAV1.0 dataset, and the validity is further verified on the RSOD dataset. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Jurassic

来源：评论

学校读者我要写书评

暂无评论

TraSculptor: Visual Analytics for Enhanced Decision-Making in Road Traffic Planning

引用

IEEE Transactions on Visualization and Computer Graphics 2025年 PP卷 PP页

作者： Deng, Zikun Liu, Yuanbang Zhu, Mingrui Xiang, Da Yu, Haiyue Su, Zicheng Lu, Qinglong Schreck, Tobias Cai, Yi South China University of Technology School of Software Engineering China South China University of Technology Ministry of Education Key Laboratory of Big Data and Intelligent Robot China Tongji University Key Laboratory of Road and Traffic Engineering of the Ministry of Education China National University of Singapore Department of Civil and Environmental Engineering Singapore Graz University of Technology Austria

The design of urban road networks significantly influences traffic conditions, underscoring the importance of informed traffic planning. Traffic planning experts rely on specialized platforms to simulate traffic systems, assessing the efficacy of the road network across various states of modifications. Nevertheless, a prevailing issue persists: many existing traffic planning platforms exhibit inefficiencies in flexibly interacting with the road network's structure and attributes and intuitively comparing multiple states during the iterative planning process. This paper introduces TraSculptor, an interactive planning decision-making system. To develop TraSculptor, we identify and address two challenges: interactive modification of road networks and intuitive comparison of multiple network states. For the first challenge, we establish flexible interactions to enable experts to easily and directly modify the road network on the map. For the second challenge, we design a comparison view with a history tree of multiple states and a road-state matrix to facilitate intuitive comparison of road network states. To evaluate TraSculptor, we provided a usage scenario where the Braess's paradox was showcased, invited experts to perform a case study on the Sioux Falls network, and collected expert feedback through interviews. © 1995-2012 IEEE.

关键词： Visual analytics

来源：评论

学校读者我要写书评

暂无评论

EDSep: An Effective Diffusion-Based Method for Speech Source Separation

EDSep: An Effective Diffusion-Based Method for Speech Source...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Dong, Jinwei Wang, Xinsheng Mao, Qirong School of Computer Science and Communication Engineering Jiangsu University China Jiangsu Engineering Research Center of Big Data Ubiquitous Perception and Intelligent Agriculture Applications China Provincial Key Laboratory of Computational Intelligence and New Technologies in Low-Altitude Digital Agriculture Zhenjiang China Audio Speech and Language Processing Group School of Computer Science Northwestern Polytechnical University Xi'an China

ISBN: (纸本)9798350368741

Generative models have attracted considerable attention for speech separation tasks, and among these, diffusion-based methods are being explored. Despite the notable success of diffusion techniques in generation tasks, their adaptation to speech separation has encountered challenges, notably slow convergence and suboptimal separation outcomes. To address these issues and enhance the efficacy of diffusion-based speech separation, we introduce EDSep, a novel single-channel method grounded in score matching via stochastic differential equation (SDE). This method enhances generative modeling for speech source separation by optimizing training and sampling efficiency. Specifically, a novel denoiser function is proposed to approximate data distributions, which obtains ideal denoiser outputs. Additionally, a stochastic sampler is carefully designed to resolve the reverse SDE during the sampling process, gradually separating speech from mixtures. Extensive experiments on databases such as WSJ0-2mix, LRS2-2mix, and VoxCeleb2-2mix demonstrate our proposed method's superior performance over existing diffusion and discriminative models, validating its efficacy. © 2025 IEEE.

关键词： diffusion score matching speech separation stochastic differential equation

来源：评论

学校读者我要写书评

暂无评论

Understanding the spatio-temporal dynamics of ecosystem services under multiple future scenarios to assess the progress of Sustainable Development Goals implementation

引用

Journal of Geographical Sciences 2025年第4期35卷 745-762页

作者： ZHANG Ze JIANG Weiguo LING Ziyan PENG Kaifeng WU Zhifeng LI Zhuo State Key Laboratory of Remote Sensing and Digital Earth Science Faculty of Geographical SciencesBeijing Normal UniversityBeijing 100875China School of Information and Control Engineering Qingdao University of TechnologyQingdao 266520ShandongChina College of Geographic and Environmental Sciences Tianjin Normal UniversityTianjin 300387China School of Geography and Remote Sensing Guangzhou UniversityGuangzhou 510006China National Marine Data and Information Service Tianjin 300171China

Ecosystem services in urban agglomerations are the environmental conditions under which human survival and development are *** assessment of ecosystem services and complex interactions can contribute positively to the achievement of the Sustainable Development Goals(SDGs)for urban ***,studies on the future contribution of multi-scenario ecosystem services to the SDGS are *** pronovel integrated modeling framework that integrates the CLUES,InVEST,SOM,and GWR approaches to address the complex relationship between ecosystem services over a long“past-present-future”time *** construct a novel ecosystem service bundle-based approach for measuring urban agglomerations progress towards achieving ecologically relevant sustainable development goals at multiple *** the future scenario,the water yield(WY),habitat quality(HQ),and soil conservation(SC)show similar spatial patterns,with comparable spatial grids,while carbon stock(CS)remains predominantly unchanged and the ecological protection scenario(EPS)improves more *** high-synergy regions are mainly distributed in bundle 4,and most of the trade-off regions appear in bundles 1 and *** the last 30 years,all but the water-related SDGs are declining in bundle 1 of the two urban agglomerations,which are 15%higher in the Guangxi Beibu Gulf(GBG)than in the Guangdong-Hong Kong-Macao Greater Bay Area(GBA).From 2020 to 2035,the three scenarios demonstrate that the optimization of the SDGs progresses most effectively under the future ecological protection scenario(EPS).In particular,bundles 3 and 4 are significantly *** critical new knowledge can be used in sustainable ecosystem management and decision-making in urban agglomerations.

关键词： Sustainable Development Goals(SDGs) ecosystem service bundle trade-offs and synergies InVEST model self-organizing mapping(SOM)algorithm urban agglomeration in south china

来源：评论

学校读者我要写书评

暂无评论

Emotional knowledge Self-Distillation in Dialogue

Emotional Knowledge Self-Distillation in Dialogue

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Jian, Zhongquan Wang, Yancheng Wu, Weichao Yao, Junfeng Wang, Meihong Wu, Qingqiang School of Informatics Xiamen University Xiamen China School of Film Xiamen University Xiamen China School of Computing and Data Science Xiamen University Malaysia Selangor Malaysia School of Mechanics and Civil Engineering China University of Mining & Technology Beijing China Xiamen Key Laboratory of Intelligent Storage and Computing School of Informatics Xiamen University Xiamen China Key Laboratory of Digital Protection and Intelligent Processing of Intangible Cultural Heritage of Fujian and Taiwan Ministry of Culture and Tourism Xiamen China

ISBN: (纸本)9798350368741

Recognizing emotions in dialogues is vital for effective human-computer interaction, yet remains a challenging task in Natural Language Processing (NLP). Previous studies in Emotion Recognition in Conversation (ERC) have primarily focused on contextual features, while overlooking the importance of emotional features in emotion recognition. To address this gap, we focus on the role of emotional features in ERC and propose a novel method, Emotional knowledge Self-Distillation (EmoKSD1), to enhance the model's emotional sensitivity. In EmoKSD, utterances are enriched with implicit 〈mask〉 tokens to represent conveyed emotions, allowing the distillation of emotional knowledge from explicit emotional tokens to implicit 〈mask〉 tokens, thereby enhancing the model's ability to perceive subtle emotions within the dialogue. Through thorough evaluations on two public ERC datasets (i.e., IEMOCAP and MELD) using proposed coarse-grained utterance distillation and fine-grained token distillation techniques, EmoKSD demonstrates superior performance compared to existing methods, highlighting the significance of emotional features in ERC. © 2025 IEEE.

关键词： Dialogue System Emotion Recognition in Conversation knowledge Distillation Natural Language Processing

来源：评论

学校读者我要写书评

暂无评论

RobNODDI: Robust NODDI Parameter Estimation with Adaptive Sampling Under Continuous Representation 15th

RobNODDI: Robust NODDI Parameter Estimation with Adaptive S...

引用

15th International Workshop on Computational Diffusion MRI, CDMRI 2024, held in conjunction with 27th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2024

作者： Xiao, Taohui Cheng, Jian Fan, Wenxin Yang, Jing Li, Cheng Dong, Enqing Wang, Shanshan School of Mechanical Electrical and Information Engineering Shandong University Weihai264209 China State Key Laboratory of Software Development Environment Beihang University Beijing China Key Laboratory of Data Science and Intelligent Computing Institute of International Innovation Beihang University Zhejiang Hangzhou China Paul C. Lauterbur Research Center for Biomedical Imaging Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong Shenzhen China Peng Cheng Laboratory Guangdong Shenzhen China

ISBN: (纸本)9783031869198

Neurite Orientation Dispersion and Density Imaging (NODDI) is an important imaging technology used to evaluate the microstructure of brain tissue, which is of great significance for the discovery and treatment of various neurological diseases. Current deep learning-based methods perform parameter estimation through diffusion magnetic resonance imaging (dMRI) with a small number of diffusion gradients. These methods speed up parameter estimation and improve accuracy. However, the diffusion directions used by most existing deep learning models during testing needs to be strictly consistent with the diffusion directions during training. This results in poor generalization and robustness of deep learning models in dMRI parameter estimation. In this work, we first verify that the parameter estimation performance of current mainstream methods will significantly decrease when the testing diffusion directions and the training diffusion directions are inconsistent. A robust NODDI parameter estimation method with adaptive sampling under continuous representation (RobNODDI) is proposed. Furthermore, long short-term memory (LSTM) units and fully connected layers are selected to learn continuous representation signals. To this end, we use a total of 100 subjects to conduct experiments based on the Human Connectome Project (HCP) dataset, of which 60 are used for training, 20 are used for validation, and 20 are used for testing. The test results indicate that RobNODDI improves the generalization performance and robustness of the deep learning model, enhancing the stability and flexibility of deep learning NODDI parameter estimatimation applications. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：