检索结果-内蒙古大学图书馆

Semantic-Aware Pseudo-Labeling for Unsupervised Meta-Learning

IEEE Transactions on Pattern Analysis and Machine Intelligence 2025年第7期47卷 5475-5488页

作者： Ouyang, Tianran Dong, Xingping Ye, Mang Du, Bo Shao, Ling Shen, Jianbing Wuhan University School of Computer Science National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan430072 China University of Chinese Academy of Sciences UCAS-Terminus AI Lab Beijing101408 China University of Macau State Key Laboratory of Internet of Things for Smart City Department of Computer and Information Science 314100 China

In unsupervised meta-learning, the clustering-based pseudo-labeling approach is an attractive framework, since it is model-agnostic, allowing it to synergize with supervised algorithms to learn from unlabeled data. However, the pseudo-labels suffer from clustering noise and semantic chaos problems, further impacting the effectiveness of meta-learning. In this paper, we analyze and optimize the pseudo-labeling process, including encoding and clustering, aiming to generate semantic-like pseudo-labels to narrow the gap between unsupervised and supervised meta-learning. First, during the encoding, we observe that the embedding space of existing methods lacks clustering-friendly properties, which is the primary reason for clustering noise. To address this issue, we minimize the inter-to-intra-class similarity ratio to generate clustering-friendly embedding features and validate our approach through comprehensive experiments. Then, during the clustering, we find that the semantic quality of pseudo-labels is not adequately controlled, resulting in semantic chaos of pseudo-labels. We propose a semantic-stability index to measure the semantic quality of pseudo-labels quantitatively. Based on this index, we propose the Semantic-aware Pseudo-label Reassignment mechanism to generate semantic-like pseudo-labels for all samples. Our approach is model-agnostic and can easily be integrated into existing supervised methods. To demonstrate its generalization ability, we integrate it into two representative algorithms: MAML and EP. The results on three main few-shot benchmarks clearly show that the proposed method achieves significant improvement compared to state-of-the-art models. Notably, our approach also outperforms the corresponding supervised method in three tasks. © 1979-2012 IEEE.

关键词： Latent semantic analysis

来源：评论

学校读者我要写书评

暂无评论

Generative AI-Driven Incentive Mechanism for Semantic Communications in RSMA Networks

引用

IEEE Transactions on Cognitive Communications and Networking 2025年第3期11卷 1708-1722页

作者： Liu, Dongqing Liu, Yongwen Zhang, Lei Hafid, Abdelhakim Khoukhi, Lyes Li, Mingyuan Liu, Yutao Dong, Yadong The First Affiliated Hospital of Zhengzhou University National Engineering Laboratory for Internet Medical Systems and Applications Zhengzhou China Zhengzhou University of Light Industry Software Engineering College Zhengzhou China Henan University of Technology College of Electrical Engineering Zhengzhou China University of Montreal Department of Computer Science and Operational Research Canada GREYC Laboratory ENSICAEN Normandie University France

This paper proposes a framework integrating Rate Splitting Multiple Access (RSMA), semantic communications, and generative AI for optimizing next-generation wireless networks. We present a unified model that combines RSMA with semantic communications to enhance both spectral efficiency and semantic fidelity. Our system model focuses on a downlink semantic communication system with a multi-antenna base station serving multiple single-antenna users. A dynamic contract-based incentive mechanism is developed to address user heterogeneity and information asymmetry in semantic RSMA scenarios. We introduce a diffusion model-based approach for joint optimization of resource allocation and contract design in RSMA systems. The semantic encoding process extracts key information, i.e., free-space detection, interest points, object attributes, and spatial relationships, from image data. A loss function is designed to train the semantic encoder and RSMA scheme, incorporating semantic extraction, partitioning, reconstruction, and task specific components. Our framework includes a multi-objective performance evaluation that considers both conventional metrics and semantic accuracy in a multi-user RSMA environment. We also define a multi-component semantic accuracy metric to assess the quality and utility of the extracted semantic information. Extensive simulation results demonstrate the superiority of our proposed framework over existing approaches in terms of system throughput, energy efficiency, and semantic fidelity across various network scenarios and user distributions. © 2015 IEEE.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Fast target-aware learning for few-shot video object segmentation

引用

science China(Information sciences) 2022年第8期65卷 71-86页

作者： Yadang CHEN Chuanyan HAO Zhi-Xin YANG Enhua WU Engineering Research Center of Digital Forensics Ministry of EducationSchool of Computer and SoftwareNanjing University of Information Science and Technology School of Education Science and Technology Nanjing University of Posts and Telecommunications State Key Laboratory of Internet of Things for Smart City Department of Electromechanical EngineeringUniversity of Macau State Key Laboratory of Computer Science Institute of SoftwareUniversity of Chinese Academy of Sciences Faculty of Science and Technology University of Macau

Few-shot video object segmentation(FSVOS) aims to segment a specific object throughout a video sequence when only the first-frame annotation is given. In this study, we develop a fast target-aware learning approach for FSVOS, where the proposed approach adapts to new video sequences from its firstframe annotation through a lightweight procedure. The proposed network comprises two models. First, the meta knowledge model learns the general semantic features for the input video image and up-samples the coarse predicted mask to the original image size. Second, the target model adapts quickly from the limited support set. Concretely, during the online inference for testing the video, we first employ fast optimization techniques to train a powerful target model by minimizing the segmentation error in the first frame and then use it to predict the subsequent frames. During the offline training, we use a bilevel-optimization strategy to mimic the full testing procedure to train the meta knowledge model across multiple video *** proposed method is trained only on an individual public video object segmentation(VOS) benchmark without additional training sets and compared favorably with state-of-the-art methods on DAVIS-2017, with a J &F overall score of 71.6%, and on YouT ubeVOS-2018, with a J &F overall score of 75.4%. Meanwhile,a high inference speed of approximately 0.13 s per frame is maintained.

关键词： video object segmentation few-shot target-aware meta knowledge bilevel-optimization

来源：评论

学校读者我要写书评

暂无评论

User story clustering in agile development:a framework and an empirical study

引用

Frontiers of computer science 2023年第6期17卷 43-59页

作者： Bo YANG Xiuyin MA Chunhui WANG Haoran GUO Huai LIU Zhi JIN School of Information Science and Technology Beijing Forestry UniversityBeijing 100083China Engineering Research Center for Forestry Oriented Intelligent Information Processing National Forestry and Grassland AdministrationBeijing 100083China School of Information Science and Technology North China University of TechnologyBeijing 100144China College of Computer Science and Technology Inner Mongolia Normal UniversityHohhot 010020China Department of Computer Science and Software Engineering Swinburne University of TechnologyHawthorn VIC 3122Australia Key Laboratory of High Confidence Software Technologies(Peking University) Ministry of EducationBeijing 100871China Institute of Software School of Computer SciencePeking UniversityBeijing 100871China

Agile development aims at rapidly developing software while embracing the continuous evolution of user requirements along the whole development *** stories are the primary means of requirements collection and elicitation in the agile development.A project can involve a large amount of user stories,which should be clustered into different groups based on their functionality’s similarity for systematic requirements analysis,effective mapping to developed features,and efficient ***,the current user story clustering is mainly conducted in a manual manner,which is time-consuming and subjective to human *** this paper,we propose a novel approach for clustering the user stories automatically on the basis of natural language ***,the sentence patterns of each component in a user story are first analysed and determined such that the critical structure in the representative tasks can be automatically extracted based on the user story *** similarity of user stories is calculated,which can be used to generate the connected graph as the basis of automatic user story *** evaluate the approach based on thirteen datasets,compared against ten baseline *** results show that our clustering approach has higher accuracy,recall rate and F1-score than these *** is demonstrated that the proposed approach can significantly improve the efficacy of user story clustering and thus enhance the overall performance of agile *** study also highlights promising research directions for more accurate requirements elicitation.

关键词： user story agile development user story mapping clustering

来源：评论

学校读者我要写书评

暂无评论

Sparse is Enough in Fine-tuning Pre-trained Large Language Models 41

Sparse is Enough in Fine-tuning Pre-trained Large Language M...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Song, Weixi Li, Zuchao Zhang, Lefei Zhao, Hai Du, Bo National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University Wuhan430072 China Hubei Luojia Laboratory Wuhan430072 China Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai200240 China

With the prevalence of pre-training-fine-tuning paradigm, how to efficiently adapt the pre-trained model to the downstream tasks has been an intriguing issue. Parameter-Efficient Fine-Tuning (PEFT) methods have been proposed for low-cost adaptation. Although PEFT has demonstrated effectiveness and been widely applied, the underlying principles are still unclear. In this paper, we adopt the PAC-Bayesian generalization error bound, viewing pre-training as a shift of prior distribution which leads to a tighter bound for generalization error. We validate this shift from the perspectives of oscillations in the loss landscape and the quasi-sparsity in gradient distribution. Based on this, we propose a gradient-based sparse finetuning algorithm, named Sparse Increment Fine-Tuning (SIFT), and validate its effectiveness on a range of tasks including the GLUE Benchmark and Instruction-tuning. The code is accessible at https://***/song-wx/SIFT. Copyright 2024 by the author(s)

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

A Robust Key Exchange and Tamper-Resistant Protocol for HAN and NAN Networks in Smart Grids

引用

IEEE Internet of Things Journal 2025年第12期12卷 21394-21403页

作者： Ayub, Muhammad Faizan Li, Xiong Mahmood, Khalid Shamshad, Salman Das, Ashok Kumar Wang, Guijuan University of Electronic Science and Technology of China School of Computer Science and Engineering Sichuan Chengdu611731 China National Yunlin University of Science and Technology Graduate School of Intelligent Data Science Yunlin Douliu64002 Taiwan The University of Lahore Department of Software Engineering Lahore54590 Pakistan International Institute of Information Technology Center for Security Theory and Algorithmic Research Hyderabad500 032 India Korea University College of Informatics Department of Computer Science and Engineering Anam-ro Seoul145 Korea Republic of Qilu University of Technology Shandong Academy of Sciences Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Jinan250353 China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Industrial Network and Information System Security Jinan250353 China

Smart Grids (SG) rely on Home Area Networks (HAN) and Neighborhood Area Networks (NAN) to ensure efficient power distribution, real-time monitoring, and seamless communication between smart devices. Despite these advantages, the use of public communication channels in HAN and NAN networks introduces critical challenges, such as vulnerability to impersonation, physical tampering, and scalability issues in resource-constrained environments. These issues compromise the stability, reliability, and security of SG environments. Existing protocols often fail to adequately address these challenges, particularly in ensuring resistance to physical tampering of supply lines and impersonation attacks. Additionally, they struggle to minimize the computation, communication, and energy costs associated with securing a resource-constrained SG environment. Therefore, we propose a robust key exchange and tamper-resistant protocol for HAN and NAN networks to address these limitations. The proposed protocol leverages the Physical Unclonable Function (PUF) mechanism to offer physical tampering resistance to smart meters. We validate our protocol formally using the Random or Real (RoR) model, which confirms its security robustness. Additionally, our informal security analysis highlights the protocol's resilience against impersonation, physical tampering attacks, etc. We analyze and compare the performance of the proposed protocol, which further demonstrates our protocol's effectiveness and security compared to competing protocols. Moreover, the proposed protocol achieves resource efficiency with 45.99% and 58.85% substantial reductions in computation overhead and energy overhead, respectively, compared to competing protocols. These results showcase the protocol's enhanced security and practicality for resource-constrained SG environments. © 2014 IEEE.

关键词： Authentication Protocol

来源：评论

学校读者我要写书评

暂无评论

A pretrained transformer model for decoding individual glucose dynamics from continuous glucose monitoring data

引用

National science Review 2025年第5期12卷 362-381页

作者： Yurun Lu Dan Liu Zhongming Liang Rui Liu Pei Chen Yitong Liu Jiachen Li Zhanying Feng Lei M.Li Bin Sheng Weiping Jia Luonan Chen Huating Li Yong Wang Center for Excellence in Mathematical Sciences National Center for Mathematics and Interdisciplinary Sciences Hua Loo-Keng Center for Mathematical Sciences Key Laboratory of Management Decision and Information System Academy of Mathematics and Systems ScienceChinese Academy of Sciences School of Mathematics University of Chinese Academy of SciencesChinese Academy of Sciences Department of Endocrinology and Metabolism Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of MedicineShanghai Diabetes InstituteShanghai Clinical Center for Diabetes Shanghai Key Laboratory of Diabetes Mellitus Key Laboratory of Systems Health Science of Zhejiang Province School of Life ScienceHangzhou Institute for Advanced StudyUniversity of Chinese Academy of Sciences BGI-Research School of Mathematics South China University of Technology Department of Statistics Department of Biomedical Data Science Bio-X ProgramStanford University Department of Computer Science and Engineering Shanghai Jiao Tong University State Key Laboratory of Cell Biology Center for Excellence in Molecular Cell Science Shanghai Institute of Biochemistry and Cell Biology Chinese Academy of Sciences Guangdong Institute of Intelligence Science and Technology Pazhou Laboratory (Huangpu)

Continuous glucose monitoring(CGM) technology has grown rapidly to track real-time blood glucose levels and trends with improved sensor accuracy. The ease of use and wide availability of CGM will facilitate safe and effective decision making for diabetes management. Here, we developed an attention-based deep learning model, CGMformer, pretrained on a well-controlled and diverse corpus of CGM data to represent individual's intrinsic metabolic state and enable clinical applications. During pretraining, CGMformer encodes glucose dynamics including glucose level, fluctuation, hyperglycemia, and hypoglycemia into latent space with self-supervised learning. It shows generalizability in imputing glucose value across five external datasets with different populations and metabolic states(MAE = 3.7 mg/d L). We then fine-tuned CGMformer towards a diverse panel of downstream tasks in the screening of diabetes and its complications using task-specific data, which demonstrated a consistently boosted predictive accuracy over direct fine-tuning on a single task(AUROC = 0.914 for type 2 diabetes(T2D) screening and 0.741 for complication screening). By learning an intrinsic representation of an individual's glucose dynamics,CGMformer classifies non-diabetic individuals into six clusters with elevated T2D risks, and identifies a specific cluster with lean body-shape but high risk of glucose metabolism disorders, which is overlooked by traditional glucose measurements. Furthermore, CGMformer achieves high accuracy in predicting an individual's postprandial glucose response with dietary modelling(Pearson correlation coefficient = 0.763)and helps personalized dietary recommendations. Overall, CGMformer pretrains a transformer neural network architecture to learn an intrinsic representation by borrowing information from a large amount of daily glucose profiles, and demonstrates predictive capabilities fine-tuned towards a broad range of downstream applications, holding promise for the ear

关键词： pretrained model diabetes continuous glucose monitoring glucose dynamics

来源：评论

学校读者我要写书评

暂无评论

Pulsed Electromagnetic Field Interaction With a Transmission Line: An Analytical Traveling-Wave Approach Based on Reciprocity

引用

IEEE Transactions on Antennas and Propagation 2025年第6期73卷 3783-3791页

作者： Stumpf, Martin Antonini, Giulio Lager, Ioan E. Ekman, Jonas The Czech Republic Lerch Laboratory of em Research Department of Radio Electronics Feec Brno University of Technology Technická Brno3082/12 616 00 Czech Republic Luleå University of Technology Department of Computer Science Electrical and Space Engineering Eislab Luleå971 87 Sweden University of l'Aquila UAq Emc Laboratory Department of Industrial and Information Engineering and Economics L'Aquila671 00 Italy Delft University of Technology Terahertz Sensing Group Faculty of Electrical Engineering Mathematics and Computer Science Delft2628 CD Netherlands

Pulsed electromagnetic (EM) field signal transfer from a general EM source distribution to a transmission line (TL) is analyzed with the aid of Lorentz's reciprocity theorem. In this fashion, the transient voltage induced by the impulsive EM source is expressed through the EM fields as radiated by the TL. These transmitted EM fields are expressed in closed form using an analytical procedure that resembles the Cagniard-DeHoop (CdH) technique. The validity of the proposed reciprocity-based methodology is verified with the aid of an alternative analytical solution describing the EM field signal transfer excited by an impulsive vertical electric dipole (VED). Illustrative numerical examples are presented. © 1963-2012 IEEE.

关键词： Time domain analysis

来源：评论

学校读者我要写书评

暂无评论

Stackelberg evolutionary games for cancer modeling and treatment 63

Stackelberg evolutionary games for cancer modeling and treat...

引用

63rd IEEE Conference on Decision and Control, CDC 2024

作者： Romano, Chiara Borri, Alessandro Benedetto, Maria Domenica Di University of l'Aquila Computer Science and Mathematics Center of Excellence for Research Dews Department of Information Engineering Italy CNR-IASI Biomathematics Laboratory National Research Council of Italy Rome Italy

ISBN: (纸本)9798350316339

Stackelberg Evolutionary Game (SEG) theory models the interacting dynamics between a rational leader and a population of evolving followers, merging classical and evolutionary game theory. Although both methods are welldeveloped individually, the potential of SEG itself has not been appropriately recognized. Thus, in this paper we propose a novel eco-evolutionary model with resource mutualism and competition and we introduce a control framework based on SEGs, to steer the eco-evolutionary dynamics of followers at will. As a case study, we consider the treatment of cancer, where tumour cells are the followers that evolve in response to changes in the tumour micro-environment (and thus on available resources) and to the medical therapy, where the physician is the leader. An interesting aspect of this approach is that the objective function can be tailored according to the goal, e.g. jointly balancing tumour size, developed resistance, and toxicity of therapy, to ensure maximum quality of life for the patient. Simulations confirm the effectiveness and the potential of the proposed approach. © 2024 IEEE.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Exploring the Predictive Capabilities of AlphaFold Using Adversarial Protein Sequences

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第7期5卷 3384-3392页

作者： Alkhouri, Ismail R. Jha, Sumit Beckus, Andre Atia, George Jha, Susmit Ewetz, Rickard Velasquez, Alvaro OrlandoFL32816 United States Michigan State University Computational Mathematics Science and Engineering Department East LansingMI48824 United States University of Michigan Electrical Engineering and Computer Science Department Ann ArborMI48109 United States Florida International University Knights Foundation School of Computing and Information Sciences MiamiFL33199 United States Air Force Research Laboratory Rome Information Directorate NY13441 United States Department Computer Science Department OrlandoFL32816 United States SRI International Computer Science Laboratory Menlo ParkCA94025 United States Electrical and Computer Engineering Department OrlandoFL32816 United States University of Colorado Computer Science Department BoulderCO80309 United States

Protein folding neural networks (PFNNs) such as AlphaFold predict remarkably accurate structures of proteins compared to other approaches. However, the robustness of such networks has heretofore not been fully explored. This is particularly relevant given the broad social implications of such technologies and the fact that biologically small perturbations to noncritical residues of a protein sequence do not typically lead to drastic changes in the protein structure. Our study demonstrates that, similar to adversarial methods in machine learning, small changes to protein sequences can result in significant differences in the predicted protein structures using AlphaFold as determined by large distance measures. Despite this, our findings using multiple protein sequences suggest that AlphaFold is able to accurately predict the domain structure and folding regions of a protein. To gauge structural differences, we employ two alignment-based measures [root-mean-square deviation (RMSD) and the global distance test (GDT) similarity], and one alignment-free measure, which is an effective graph-based structure representation (GraSR) method. We prove that the problem of minimally perturbing protein sequences is NP-complete. Based on the well-established block substitution matrices (BLOSUM62) sequence alignment scoring matrix, we generate adversarial sequences. In our experimental evaluation, we consider 111 proteins (including 29 COVID-19 sequences) in the Universal Protein resource (UniProt), a central resource for protein data. Our findings suggest that, despite the high RMSD values returned by AlphaFold, it is capable of handling the BLOSUM adversarial sequences considered in our analysis, as evidenced by the preservation of the folded regions and the GraSR results. © 2020 IEEE.

关键词： Neural networks

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：