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 PP卷 PP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 Beijing China University of Macau State Key Laboratory of Internet of Things for Smart City Department of Computer and Information Science 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. Firstly, 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. The source code will be made publicly available. ©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

MRLATO: An Adaptive Task Offloading Mechanism Based on Meta Reinforcement Learning in Edge Computing Environment

引用

IEEE Transactions on Vehicular Technology 2025年

作者： Zhang, Peiying Liu, Jiamin Guizani, Maher Wang, Jian Kumar, Neeraj Tan, Lizhuang Qingdao Institute of Software College of Computer Science and Technology Qingdao266580 China Shandong Key Laboratory of Intelligent Oil & Gas Industrial Software Qingdao266580 China Ministry of Education Qilu University of Technology Shandong Academy of Sciences Key Laboratory of Computing Power Network and Information Security Jinan250014 China University of Texas Arlington Computer Science and Engineering Department TX United States College of Science Qingdao266580 China Thapar University Department of Computer Science and Engineering Patiala147004 India Jinan250014 China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Computing Power Internet and Service Computing Jinan250014 China

Traditional cloud computing models struggle to meet the requirements of latency-sensitive applications when processing large amounts of data. As a solution, Multi-access Edge Computing (MEC) extends computing resources to the edge of the network to reduce processing delays and improve user experience. However, in dynamically changing edge computing environments, effective decision making on whether to offload tasks to edge servers remains a core challenge. For this purpose, we propose MRLATO, an adaptive task offloading mechanism based on Meta Reinforcement Learning (MRL), which exploits a large amount of a priori knowledge of different tasks to achieve fast adaptation. The task offloading process is modelled as multiple Markov Decision Processes (MDPs) and solved using a Sequence to Sequence (Seq2Seq) neural network integrating multi-head attention and recursive task sequencing. It is shown by the experimental results that the proposed method has the lowest latency in all experimental settings and the convergence efficiency is improved by 14.06% compared to the traditional Deep Reinforcement Learning (DRL) algorithm. This research fully demonstrates the significant benefits of the deep integration of MRL with the edge computing domain, providing new optimisation ideas for task offloading decisions. © 1967-2012 IEEE.

关键词： Computation offloading

来源：评论

学校读者我要写书评

暂无评论

FM-FCN:A Neural Network with Filtering Modules for Accurate Vital Signs Extraction

引用

Research 2025年第1期2024卷 92-106页

作者： Fangfang Zhu Qichao Niu Xiang Li Qi Zhao Honghong Su Jianwei Shuai Department of Physics and Fujian Provincial Key Laboratory for Soft Functional Materials ResearchXiamen UniversityXiamen 361005China National Institute for Data Science in Health and Medicine and State Key Laboratory of Cellular Stress BiologyInnovation Center for Cell Signaling NetworkXiamen UniversityXiamen 361005China Vitalsilicon Technology Co.Ltd. JiaxingZhejiang 314006China School of Computer Science and Software Engineering University of Science and Technology LiaoningAnshan 114051China Yangtze Delta Region Institute of Tsinghua University ZhejiangJiaxing 314006China Wenzhou Institute University of Chinese Academy of SciencesWenzhou 325001China Oujiang Laboratory (Zhejiang Lab for Regenerative Medicine Vision and Brain Health)Wenzhou 325001China

Neural networks excel at capturing local spatial patterns through convolutional modules,but they may struggle to identify and effectively utilize the morphological and amplitude periodic nature of physiological *** this work,we propose a novel network named filtering module fully convolutional network(FM-FCN),which fuses traditional filtering techniques with neural networks to amplify physiological signals and suppress ***,instead of using a fully connected layer,we use an FCN to preserve the time-dimensional correlation information of physiological signals,enabling multiple cycles of signals in the network and providing a basis for signal ***,we introduce the FM as a network module that adapts to eliminate unwanted interference,leveraging the structure of the *** approach builds a bridge between deep learning and signal processing ***,we evaluate the performance of FM-FCN using remote *** results demonstrate that FM-FCN outperforms the second-ranked method in terms of both blood volume pulse(BVP)signal and heart rate(HR)*** substantially improves the quality of BVP waveform reconstruction,with a decrease of 20.23%in mean absolute error(MAE)and an increase of 79.95%in signal-to-noise ratio(SNR).Regarding HR estimation accuracy,FM-FCN achieves a decrease of 35.85%in MAE,29.65%in error standard deviation,and 32.88%decrease in 95%limits of agreement width,meeting clinical standards for HR accuracy *** results highlight its potential in improving the accuracy and reliability of vital sign measurement through high-quality BVP signal *** codes and datasets are available online at https://***/zhaoqi106/FM-FCN.

关键词： physiological signalsin filtering module fully convolutional network fm fcn which vital signs extraction amplify physiological signals convolutional modulesbut neural networks filtering module capturing local spatial patterns

来源：评论

学校读者我要写书评

暂无评论

OpenLS-DGF: An Adaptive Open-Source Dataset Generation Framework for Machine Learning Tasks in Logic Synthesis

引用

IEEE Transactions on computer-Aided Design of Integrated Circuits and Systems 2025年

作者： Ni, Liwei Wang, Rui Liu, Miao Meng, Xingyu Lin, Xiaoze Liu, Junfeng Luo, Guojie Chu, Zhufei Qian, Weikang Yang, Xiaoyan Xie, Biwei Li, Xingquan Li, Huawei Chinese Academy of Sciences State Key Lab of Processors Institute of Computing Technology Beijing100190 China Pengcheng Laboratory Shenzhen518055 China University of Chinese Academy of Sciences Beijing101408 China Shenzhen University College of Computer Science and Software Engineering Shenzhen518060 China University of Chinese Academy of Sciences School of Computer Science and Technology Beijing100049 China Peking University School of Computer Science Center for Energy-Efficient Computing and Applications Beijing100871 China Ninbo University Faculty of Electrical Engineering and Computer Science Ninbo315211 China Shanghai Jiao Tong University University of Michigan-Shanghai Jiao Tong University Joint Institute MoE Key Laboratory of Artificial Intelligence Shanghai200240 China Hangzhou Dianzi University School of Electronics and Information Engineering Hangzhou311121 China

This paper introduces OpenLS-DGF, an adaptive logic synthesis dataset generation framework, to enhance machine learning (ML) applications within the logic synthesis process. Previous dataset generation flows were tailored for specific tasks or lacked integrated ML capabilities. While OpenLS-DGF supports various machine learning tasks by encapsulating the three fundamental steps of logic synthesis: Boolean representation, logic optimization, and technology mapping. It preserves the original information in both Verilog and ML-friendly GraphML formats. The Verilog files offer semi-customizable capabilities, enabling researchers to insert additional steps and incrementally refine the generated dataset. Furthermore, OpenLS-DGF includes an adaptive circuit engine that facilitates the final dataset management and downstream tasks. The generated OpenLS-D-v1 dataset comprises 46 combinational designs from established benchmarks, totaling over 966,000 Boolean circuits. OpenLS-D-v1 supports integrating new data features, making it more versatile for new tasks. This paper demonstrates the versatility of OpenLS-D-v1 through four distinct downstream tasks: circuit classification, circuit ranking, quality of results (QoR) prediction, and probability prediction. Each task is chosen to represent essential steps of logic synthesis, and the experimental results show the generated dataset from OpenLSDGF achieves prominent diversity and applicability. The source code and datasets are available at https://***/Logic-Factory/ACE/blob/master/OpenLS-DGF. © 1982-2012 IEEE.

关键词： Information management

来源：评论

学校读者我要写书评

暂无评论

MindScore: quantifying human preference for text-to-image generation through multi-view lens

引用

science China(Information sciences) 2025年第06期 72-85页

作者： Yiqi TONG Jiarui ZHANG Shaohang WEI Wei GUO Fuzhen ZHUANG Deqing WANG Xi YANG Richeng XUAN School of Artificial Intelligence Beihang University School of Computer Science and Engineering Beihang University Department of Computer Science and Engineering Shanghai Jiao Tong University School of Computer Science Peking University State Key Laboratory of Complex & Critical Software Environment Beihang University Beijing Academy of Artificial Intelligence

Understanding and quantifying the capabilities of foundation models, particularly in text-to-image(T2I) generation, is crucial for verifying their alignment with human expectations and practical requirements. However, evaluating T2I foundation models presents significant challenges due to the complex, multi-dimensional psychological factors that influence human preferences for generated images. In this work, we propose MindScore, a multi-view framework for assessing the generation capacity of T2I models through the lens of human preference. Specifically, MindScore decomposes the evaluation into four complementary modules that align with human cognitive processing of images: matching, faithfulness, quality,and realness. The matching module quantifies the semantic alignment between generated images and prompt text, while the faithfulness module measures how accurately the images reflect specific prompt details. Furthermore, we incorporate quality and realness modules to capture deeper psychological preferences, recognizing that unpleasant or distorted images often trigger adverse human responses. Extensive experiments on three T2I datasets with human preference annotations clearly validate the superiority of our proposed MindScore over various state-of-the-art baselines. Our case studies further reveal that MindScore offers valuable insights into T2I generation from a human-centric perspective.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Flat U-Net: An Efficient Ultralightweight Model for Solar Filament Segmentation in Full-disk Hα Images

arXiv

引用

arXiv 2025年

作者： Zhu, GaoFei Lin, GangHua Yang, Xiao Zeng, Cheng School of Computer and Software Engineering Xihua University Chengdu610039 China National Astronomical Observatories Chinese Academy of Sciences Beijing100101 China State Key Laboratory of Solar Activity and Space Weather National Space Science Center Chinese Academy of Sciences Beijing100190 China

Solar filaments are one of the most prominent features observed on the Sun, and their evolutions are closely related to various solar activities, such as flares and coronal mass ejections. Real-time automated identification of solar filaments is the most effective approach to managing large volumes of data. Existing models of filament identification are characterized by large parameter sizes and high computational costs, which limit their future applications in highly integrated and intelligent ground-based and space-borne observation devices. Consequently, the design of more lightweight models will facilitate the advancement of intelligent observation equipment. In this study, we introduce Flat U-Net, a novel and highly efficient ultralightweight model that incorporates simplified channel attention (SCA) and channel self-attention (CSA) convolutional blocks for the segmentation of solar filaments in full-disk Hα images. Feature information from each network layer is fully extracted to reconstruct interchannel feature representations. Each block effectively optimizes the channel features from the previous layer, significantly reducing parameters. The network architecture presents an elegant flattening, improving its efficiency, and simplifying the overall design. Experimental validation demonstrates that a model composed of pure SCAs achieves a precision of approximately 0.93, with dice similarity coefficient (DSC) and recall rates of 0.76 and 0.64, respectively, significantly outperforming the classical U-Net. Introducing a certain number of CSA blocks improves the DSC and recall rates to 0.82 and 0.74, respectively, which demonstrates a pronounced advantage, particularly concerning model weight size and detection effectiveness. The data set, models, and code are available as open-source resources. Copyright © 2025, The Authors. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Climate Downscaling Using Neural Operator: Spatiotemporal Multimodal Fusion Operator with State-Query Coupled Kernel

Climate Downscaling Using Neural Operator: Spatiotemporal Mu...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Haodi Zhang Yichi Wang Yifan Jian Jiahui Jiang Zhaohai Bai Lin Ma College of Computer Science and Software Engineering Shenzhen University Center for Agricultural Resources Research Institute of Genetic and Developmental Biology Chinese Academy of Sciences State Key Laboratory of Pollution Control and Resource Reuse School of the Environment Nanjing University

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Climate downscaling is crucial for detailed small- scale analysis and for acquiring climate data in regions without weather stations. Operator learning has proven potential for this task. However, several challenges remain in operator learning, such as multimodal fusion, spatiotemporal fusion and input state and query adaptation. To address these challenges, we propose a Spatiotemporal Multimodal Fusion Operator with a State- Query Coupled Kernel (SMCK). This framework includes a latent space fusion encoder that encodes climate variables using position-wise multihead attention for multimodal fusion and integrates historical information to generate robust and precise representation. Additionally, we introduce a state-query coupled kernel that combines radial basis functions and discrete fourier encoding to enhance query location representation, while also adapting to the state to obtain the coupled kernel. Extensive experiments demonstrate that our method achieves state-of-the-art performance and provides strong support for climate downscaling and the planning of climate-related strategies.

关键词： Couplings Signal processing Encoding Robustness Spatiotemporal phenomena Planning Kernel Speech processing Forecasting Meteorology

来源：评论

学校读者我要写书评

暂无评论

Searching for an Accurate Robot Calibration via Improved Levenberg–Marquardt and Radial Basis Function System

引用

Journal of Field Robotics 2025年第0期

作者： Li, Zhibin Deng, Xun Chen, Tinghui Yang, Yuhang Chen, Linlin Yang, Xiwen Hu, Zhenzhen Hu, Lun Hu, Pengwei Li, Shuai Luo, Xin Xinjiang Technical Institute of Physics & Chemistry Chinese Academy of Sciences Urumqi China School of Software Engineering Chengdu University of Information Technology Chengdu China Dazhou Key Laboratory of Government Data Security Sichuan University of Arts and Science Dazhou China School of Software Engineering Xinjiang University Urumqi China Faculty of Information Technology and Electrical Engineering University of Oulu Oulu Finland VTT-Technical Research Centre of Finland Oulu Finland College of Computer and Information Science Southwest University Chongqing China

Robots are frequently utilized in manufacturing, aviation, and other industries, which enhance industrial production efficiency and quality. Specifically, robots perform high-precision tasks like welding, assembly and material handling, which reduce the intensity of manual labor in factories. In the logistics field, robots automatically sort and deliver goods, thereby speeding up supply chain operations. However, the prolonged operation of robots suffers from a decline in positioning accuracy, which makes them unable to satisfy task requirements. To address this challenging issue, this study designs an efficient calibration system integrating the Levenberg–Marquardt algorithm with fuzzy proportion integration differentiation controller and radial basis function neural network. The innovations of this method include: (1) integrating the fuzzy proportion integration differentiation controller into the updating rules of Levenberg–Marquardt algorithm, which further enhances the identification performance of kinematic errors;(2) adopting the radial basis function neural network to handle the robot dynamic errors, which addresses the complexity of dynamic error sources. Extensively experimental robot positioning points are gathered on an HSR JR680 robot, and then experimental validations are conducted by using the designed calibration system. The experiments indicate that the developed algorithm outperforms these existing advanced algorithms. © 2025 Wiley Periodicals LLC.

关键词： Radial basis function networks

来源：评论

学校读者我要写书评

暂无评论

Machine Learning for Traffic-Driven Services Provisioning in Optical Networks

Machine Learning for Traffic-Driven Services Provisioning in...

引用

Advanced Computing Technologies (ICoACT), International Conference on

作者： L. Sanker Magadevan N. Shanmugalakshmi M. Giridhar M. Balaji R. Saravanakumar S. Jeya Lakshmi Master of Computer Application Karpagam Academy of Higher Education Coimbatore Tamil Nadu India Department of Computer Application Karpagam Academy of Higher Education Coimbatore Tamil Nadu India Department of Computer Science and Engineering P.S.R Engineering College Sivakasi Tamil Nadu India Department of Artificial Intelligence & Data Science Vel Tech High Tech Dr. Rangarajan and Dr. Sakunthala Engineering College Chennai Tamil Nadu India Iconix Software Solution Tirunelveli Tamil Nadu India

ISBN: (数字)9798331509859

ISBN: (纸本)9798331509866

The increasing demand for high-capacity and dynamic services in optical networks necessitates intelligent and adaptive provisioning mechanisms. This study investigates the application of machine learning (ML) techniques for traffic-driven service provisioning in optical networks, addressing challenges such as traffic prediction, resource allocation, and fault management. The proposed framework employs supervised and unsupervised learning models, including Long Short-Term Memory (LSTM) networks, Random Forests, and K-Means clustering, to analyze traffic patterns and optimize service deployment. The system leverages real-time and historical traffic data to predict demand, enabling proactive resource allocation and dynamic wavelength assignment. Furthermore, reinforcement learning is explored to automate decision- making for adaptive routing and spectrum management under varying network conditions. Performance evaluation on simulated and real-world datasets demonstrates that ML-based approaches significantly improve network efficiency, reduce latency, and enhance the quality of service (QoS). This research highlights the potential of machine learning in revolutionizing service provisioning in optical networks, offering scalable, data-driven solutions to meet the growing demands of next-generation communication systems.

关键词： Reviews Buildings Machine learning Quality of service Optical fiber networks Elastic optical networks Resource management Monitoring Long short term memory Random forests

来源：评论

学校读者我要写书评

暂无评论

Privacy-Aware Edge Computation Offloading With Federated Learning in Healthcare Consumer Electronics System

引用

IEEE Transactions on Consumer Electronics 2025年

作者： Fei, Yiming Fang, Hao Yan, Zheng Qi, Lianyong Bilal, Muhammad Li, Yang Xu, Xiaolong Zhou, Xiaokang Nanjing University of Information Science and Technology School of Software Nanjing210044 China NanJing University of Information Science and Technology Reading Academy Nanjing210044 China Xidian University School of Cyber Engineering Shaanxi Xi'an710026 China College of computer science and technology Qingdao China Lancaster University School of Computing and Communications United Kingdom Shihezi University College of Mechanical and Electrical Engineering Shihezi832003 China Kansai University Faculty of Business and Data Science Japan

The continuous iteration of consumer electronics has significantly promoted the development of medical devices, which has enabled the collection of large amounts of heterogeneous medical data. These data are offloaded from local devices to cloud servers for processing using traditional methods, which results in high transmission latency and the risk of privacy leakage. Additionally, researchers have employed federated learning to protect data privacy, but this approach can lead to a straggler effect due to the limited computational and communication resources of terminal devices. To address these issues, a computation offloading framework is designed to optimize task and resource allocation, addressing multi-optimization problems and mitigating the straggler effect in federated learning. Moreover, a novel computation offloading method within a federated learning framework assisted by edge computing, named DRWB, is proposed. Specifically, we develop a deep reinforcement learning-based approach to transfer lagging training tasks to idle edge servers, enhancing data processing speed, minimizing transmission delays, and protecting user privacy. Extensive experimental results demonstrate that the DRWB method outperforms baseline methods, showcasing superior performance in handling heterogeneous medical data tasks. © 1975-2011 IEEE.

关键词： Electronic health record

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

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

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：