检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Leng, Jiaqi Zheng, Yufan Jia, Zhiyuan Fan, Lei Zhao, Chaoyue Peng, Yuxiang Wu, Xiaodi Simons Institute for the Theory of Computing University of California Berkeley United States Department of Mathematics University of California Berkeley United States Department of Computer Science University of Maryland College Park United States Joint Center for Quantum Information and Computer Science University of Maryland United States Department of Applied Mathematics University of Washington United States Department of Engineering Technology University of Houston United States Department of Electrical and Computer Engineering University of Houston United States Department of Industrial and Systems Engineering University of Washington United States Artephi Computing Inc United States

Non-smooth optimization models play a fundamental role in various disciplines, including engineering, science, management, and finance. However, classical algorithms for solving such models often struggle with convergence speed, scalability, and parameter tuning, particularly in high-dimensional and non-convex settings. In this paper, we explore how quantum mechanics can be leveraged to overcome these limitations. Specifically, we investigate the theoretical properties of the Quantum Hamiltonian Descent (QHD) algorithm for non-smooth optimization in both continuous and discrete time. First, we propose continuous-time variants of the general QHD algorithm and establish their global convergence and convergence rate for non-smooth convex and strongly convex problems through a novel Lyapunov function design. Furthermore, we prove the finite-time global convergence of continuous-time QHD for non-smooth non-convex problems under mild conditions (i.e., locally Lipschitz). In addition, we propose discrete-time QHD, a fully digitized implementation of QHD via operator splitting (i.e., product formula). We find that discrete-time QHD exhibits similar convergence properties even with large time steps. Finally, numerical experiments validate our theoretical findings and demonstrate the computational advantages of QHD over classical non-smooth non-convex optimization algorithms. Copyright © 2025, The Authors. All rights reserved.

关键词： Continuous time systems

来源：评论

学校读者我要写书评

暂无评论

Fe2O3-Functionalized MoS2 Nanostructure Sensor for High-Sensitivity and Low-Level SO2 Detection

引用

ACS Sensors 2025年

作者： Gond, Rahul Barala, Suraj Shukla, Prajjwal Bassi, Gaurav Kumar, Suresh Kumar, Mukesh Kumar, Mahesh Rawat, Brajesh Department of Electrical Engineering Indian Institute of Technology Ropar Punjab Rupnagar140001 India IDRP-Space Science and Technology Indian Institute of Technology Jodhpur Rajasthan Jodhpur342030 India Department of Physics Indian Institute of Technology Ropar Punjab Rupnagar140001 India Department of Physics Indian Institute of Technology Jodhpur Rajasthan Jodhpur342030 India Department of Electrical Engineering Indian Institute of Technology Jodhpur Rajasthan Jodhpur342030 India Department of Cybernetics Nanotechnology and Data Processing Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Akademicka 16 Gliwice44-100 Poland

Real-time sulfur dioxide (SO2) monitoring is essential to mitigate its severe health and environmental impacts while ensuring compliance with industrial safety and emission regulations. Two-dimensional MoS2 stands out as a promising material for developing low-temperature-operated gas sensors due to its exceptionally high surface-to-volume ratio and ease of surface functionalization. However, monitoring the SO2 level faces challenges, including limited selectivity, sensitivity, and detection range, with high operating temperatures (200-600 °C) or external light source requirements. To address these issues, we present a highly sensitive SO2 sensor based on Fe2O3 nanoparticle-functionalized vertically aligned MoS2 nanostructure material, which is fabricated using a scalable sputtering process. The Fe2O3-MoS2 sensor exhibits a broad detection range from 100 ppb to 100 ppm, with a theoretical detection limit of around 22.8 ppb. When exposed to 5 ppm of SO2, the sensor achieves a response of around 32.2%, with response and recovery times of approximately 104 and 141 s, respectively. The fabricated sensor demonstrated impressive sensitivity (4.9%/ppm) for SO2 concentration in the range of 0.1 to 5 ppm, coupled with excellent reproducibility and stability at 150 °C. This enhanced performance is attributed to the catalytic effect of Fe2O3 and modulation of the heterojunction barrier at the interface. This study introduces a highly scalable, reliable, and stable Fe2O3-MoS2 sensor, which paves the way for developing energy-efficient and miniaturized SO2 sensors. © 2025 American Chemical Society.

关键词： Layered semiconductors

来源：评论

学校读者我要写书评

暂无评论

NOMANet: A Graph Neural Network Enabled Power Allocation Scheme for NOMA

arXiv

引用

arXiv 2025年

作者： Hou, Yipu Lu, Yang Chen, Wei Ai, Bo Niyato, Dusit Ding, Zhiguo State Key Laboratory of Advanced Rail Autonomous Operation China School of Computer Science and Technology Beijing Jiaotong University Beijing100044 China School of Electronics and Information Engineering Beijing Jiaotong University Beijing100044 China College of Computing and Data Science Nanyang Technological University 639798 Singapore Department of Electrical Engineering and Computer Science Khalifa University Abu Dhabi127788 United Arab Emirates

This paper proposes a graph neural network (GNN) enabled power allocation scheme for non-orthogonal multiple access (NOMA) networks. In particular, a downlink scenario with one base station serving multiple users over several subchannels is considered, where the number of subchannels is less than the number of users, and thus, some users have to share a subchannel via NOMA. Our goal is to maximize the system energy efficiency subject to the rate requirement of each user and the overall budget. We propose a deep learning based approach termed NOMA net (NOMANet) to address the considered problem. Particularly, NOMANet is GNN-based, which maps channel state information to the desired power allocation scheme for all subchannels. The multi-head attention and the residual/dense connection are adopted to enhance the feature extraction. The output of NOMANet is guaranteed to be feasible via the customized activation function and the penalty method. Numerical results show that NOMANet trained unsupervised achieves performance close to that of the successive convex approximation method but with a faster inference speed by about 700 times. Besides, NOMANet is featured by its scalability to both users and subchannels. © 2025, CC BY.

关键词： Channel state information

来源：评论

学校读者我要写书评

暂无评论

Spiking Neural Networks for Temporal Processing: Status Quo and Future Prospects

arXiv

引用

arXiv 2025年

作者： Ma, Chenxiang Chen, Xinyi Li, Yanchen Yang, Qu Wu, Yujie Li, Guoqi Pan, Gang Tang, Huajin Tan, Kay Chen Wu, Jibin Department of Data Science and Artificial Intelligence The Hong Kong Polytechnic University Hong Kong Department of Computing The Hong Kong Polytechnic University Hong Kong Department of Electrical and Computer Engineering National University of Singapore 119077 Singapore Institute of Automation Chinese Academy of Sciences Beijing100045 China State Key Laboratory of Brain-Machine Intelligence College of Computer Science and Technology MOE Frontier Science Center for Brain Science and Brain-Machine Integration Zhejiang University Hangzhou310027 China

Temporal processing is fundamental for both biological and artificial intelligence systems, as it enables the comprehension of dynamic environments and facilitates timely responses. Spiking Neural Networks (SNNs) excel in handling such data with high efficiency, owing to their rich neuronal dynamics and sparse activity patterns. Given the recent surge in the development of SNNs, there is an urgent need for a comprehensive evaluation of their temporal processing capabilities. In this paper, we first conduct an in-depth assessment of commonly used neuromorphic benchmarks, revealing critical limitations in their ability to evaluate the temporal processing capabilities of SNNs. To bridge this gap, we further introduce a benchmark suite consisting of three temporal processing tasks characterized by rich temporal dynamics across multiple timescales. Utilizing this benchmark suite, we perform a thorough evaluation of recently introduced SNN approaches to elucidate the current status of SNNs in temporal processing. Our findings indicate significant advancements in recently developed spiking neuron models and neural architectures regarding their temporal processing capabilities, while also highlighting a performance gap in handling long-range dependencies when compared to state-of-the-art non-spiking models. Finally, we discuss the key challenges and outline potential avenues for future research. Copyright © 2025, The Authors. All rights reserved.

关键词： Spatio-temporal data

来源：评论

学校读者我要写书评

暂无评论

Efficient COVID-19 detection using data mining algorithms: a comparison of basic and hybrid approaches

引用

Soft computing 2025年第3期29卷 1437-1451页

作者： Saidi, Mohammad Gheibi, Mohammad Ghazikhani, Adel Lotfata, Aynaz Chahkandi, Benyamin Familsamavati, Sajad Behzadian, Kourosh Department of Computer Science Imam Reza International University Mashhad Iran MS of Software Engineering Information and Communication Technology Unit Mashhad Municipality Fire and Safety Services Organization Mashhad Iran Big Data Lab Imam Reza International University Mashhad Iran Institute for Nanomaterials Advanced Technologies and Innovation Technical University of Liberec Liberec46117 Czech Republic Tallinn10615 Estonia Department of Pathology Microbiology and Immunology School of Veterinary Medicine University of California Davis95616 United States Department of Urban and Regional Planning San Jose State University San Jose95192 United States Faculty of Civil and Environmental Engineering Gdansk University of Technology Narutowicza Street 11/12 80-233 Gdansk Poland Tecnologico de Monterrey Escuela de Ingeniería y Ciencias Monterrey Mexico School of Computing and Engineering University of West London London United Kingdom

Accurate and efficient diagnosis of COVID-19 remains a significant challenge due to the limitations of current detection methods, such as blood tests and chest scans, which can be time-consuming and error-prone. This study aims to compare the performance of basic and hybrid data mining algorithms in diagnosing COVID-19, using blood test results and clinical information to identify the most effective approach. A dataset of 200 records from suspected and infected COVID-19 patients, with 23 characteristics and one diagnostic class, was analysed. Nine data mining algorithms were tested: four basic algorithms (Naive Bayes, Support Vector Machine, Decision Tree, K-Nearest Neighbor) and five hybrid algorithms (Random Forest, AdaBoost, Majority Voting, XGBoost, Bagging). The study also integrated Response Surface Methodology (RSM) and Adaptive-Network-based Fuzzy Inference System (ANFIS) to enhance model performance. The Bagging algorithm demonstrated superior performance with an accuracy of 88%, sensitivity of 74%, and F-criterion of 78%. The integration of RSM and ANFIS further showed that a smart model could be developed for efficient pandemic crisis management, achieving up to 100% accuracy when considering key factors like AST, Albumin, and CRP. The findings suggest that Bagging and hybrid data mining algorithms can significantly improve COVID-19 detection, reducing time and errors in identifying exposed individuals. The study highlights the potential of combining machine learning techniques with RSM-ANFIS models for effective pandemic management and decision-making in medical settings. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

IoT-Enabled Smart Irrigation System with ABiLSTM for Optimized Water Management and Crop Yields

IoT-Enabled Smart Irrigation System with ABiLSTM for Optimiz...

引用

Integrated Circuits and Communication Systems (ICICACS), IEEE International Conference on

作者： Abhay Chaturvedi Praveena Rao Venkata Maruti Prasad S Malini T S. Prabagar K. Bharathi Department of Electronics and Communication Engineering GLA University Mathurai India Department of Civil Engineering Institute of Aeronautical Engineering Hyderabad India Department of BS & H Aditya Institute of Technology & Management Tekkali India Department of Electrical and Electronics Engineering Sri Krishna College of Engineering and Technology Coimbatore India Department of Computer Science and Engineering School of Advanced Computing Alliance University Bangalore India Department of Agricultural Engineering Erode Sengunthar Engineering College Perundurai India

ISBN: (数字)9798331508456

ISBN: (纸本)9798331508463

An efficient management of water resources in agriculture is now crucial in light of the growing global population and increased demand for water. Innovative solutions are needed to improve water efficiency since conventional irrigation methods are not sustainable anymore. Through the use of precision agriculture and the Internet of Things (IoT), smart irrigation has revolutionised water management, leading to increased crop yields. In order to improve the efficiency of decision-making, this research presents an Internet of Things (IoT) smart irrigation system that uses SA-BiLSTM (Self-Attention Bidirectional Long Short-Term Memory). Using Min-Max and Z-score normalisation approaches for feature scaling, the methodology includes data preparation, feature selection, and model training. Methods based on ranking, such as the Chi-Square Test, Pearson Correlation Coefficient (PCC), and Mutual Information (MI), were used to prioritise feature selection. Classification and prediction testing showed that the proposed SA-BiLSTM model outperformed state-of-the-art models like HGAT and Linear SVM, with an astounding 95.63% accuracy. Improved resource allocation and irrigation effectiveness, leading to increased agricultural output, are highlighted by the results, which highlight the effectiveness of advanced machine learning approaches in precision agriculture.

关键词： Support vector machines Precision agriculture Irrigation Accuracy Predictive models Feature extraction Internet of Things Water resources Long short term memory Mutual information

来源：评论

学校读者我要写书评

暂无评论

Improving community-based electricity markets regulation: A holistic multi-objective optimization framework

引用

Renewable and Sustainable Energy Reviews 2025年 215卷

作者： Costa, V.B.F. Soares, T. Bitencourt, L. Dias, B.H. Deccache, E. Silva, B.M.A. Bonatto, B. Filho, W.F. Faria, A.S. Institute of Electrical Systems and Energy Federal University of Itajuba MG Itajuba37500-903 Brazil Centre for Power and Energy Systems Institute of Systems and Computer Engineering Technology and Science Porto Portugal Electrical Engineering Department Federal Fluminense University RJ Niterói24210-346 Brazil Electrical Energy Department Federal University of Juiz de Fora MG Juiz de Fora36036-900 Brazil Faculty of Engineering of the University of Porto Porto Portugal Faculty of Electrical and Computing Engineering State University of Campinas SP Campinas13083-852 Brazil

Community-based electricity markets, which are defined as groups of members that share common interests in renewable distributed generation, allow prosumers to embrace more active roles by opening up several opportunities for trading electricity. At the same time, such markets may favor conventional consumers by allowing them to choose cheaper electricity providers. Due to trends in power sector modernization, community-based electricity markets are of great research interest, and there are already some associated models. However, there is a research gap in searching for integrated and holistic approaches that go beyond economic aspects, consider social and environmental aspects, and assume the balanced co-existence of community-based and conventional markets. This work fills this critical research gap by adapting/applying the optimized tariff model, Bass diffusion model, life cycle assessment, and multi-objective optimization to the context of community-based markets. Results indicate that favoring conventional markets in the short term and community-based markets in the medium term is beneficial. Moreover, regulated tariffs should increase slightly in the short/medium-term to accommodate DG growth. Additionally, community-based markets can decrease electricity expenses by around 13.6 % considering the market participants. Thus, such markets can be significantly beneficial in mitigating energy poverty. © 2025 Elsevier Ltd

关键词： Community-based electricity markets Distributed renewable generation Global warming potential Multi-objective optimization Regulatory framework Socioeconomic welfare

来源：评论

学校读者我要写书评

暂无评论

Tutorial on Using Machine Learning and Deep Learning Models for Mental Illness Detection

arXiv

引用

arXiv 2025年

作者： Zhang, Yeyubei Wang, Zhongyan Ding, Zhanyi Tian, Yexin Dai, Jianglai Shen, Xiaorui Liu, Yunchong Cao, Yuchen University of Pennsylvania School of Engineering and Applied Science United States New York University Center for Data Science United States Georgia Institute of Technology College of Computing United States University of California Department of EECS Berkeley United States Northeastern University Khoury College of Computer Science United States

Social media has become an important source for understanding mental health, providing researchers a way to detect conditions like depression from user-generated posts. This tutorial provides practical guidance to address common challenges in applying machine learning and deep learning methods for mental health detection on these platforms. It focuses on strategies for working with diverse datasets, improving text preprocessing, and addressing issues such as imbalanced data and model evaluation. Real-world examples and step-by-step instructions demonstrate how to apply these techniques effectively, with an emphasis on transparency, reproducibility, and ethical considerations. By sharing these approaches, this tutorial aims to help researchers build more reliable and widely applicable models for mental health research, contributing to better tools for early detection and intervention. © 2025, CC BY.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Efficient or Powerful? Trade-offs Between Machine Learning and Deep Learning for Mental Illness Detection on Social Media

arXiv

引用

arXiv 2025年

作者： Ding, Zhanyi Wang, Zhongyan Zhang, Yeyubei Cao, Yuchen Liu, Yunchong Shen, Xiaorui Tian, Yexin Dai, Jianglai University of Pennsylvania School of Engineering and Applied Science United States New York University Center for Data Science United States Georgia Institute of Technology College of Computing United States University of California Berkeley Department of EECS United States Northeastern University Khoury College of Computer Science United States

Social media platforms provide valuable insights into mental health trends by capturing user-generated discussions on conditions such as depression, anxiety, and suicidal ideation. Machine learning (ML) and deep learning (DL) models have been increasingly applied to classify mental health conditions from textual data, but selecting the most effective model involves trade-offs in accuracy, interpretability, and computational efficiency. This study evaluates multiple ML models, including logistic regression, random forest, and LightGBM, alongside deep learning architectures such as ALBERT and Gated Recurrent Units (GRUs), for both binary and multi-class classification of mental health conditions. Our findings indicate that ML and DL models achieve comparable classification performance on medium-sized datasets, with ML models offering greater interpretability through variable importance scores, while DL models are more robust to complex linguistic patterns. Additionally, ML models require explicit feature engineering, whereas DL models learn hierarchical representations directly from text. Logistic regression provides the advantage of capturing both positive and negative associations between features and mental health conditions, whereas tree-based models prioritize decision-making power through split-based feature selection. This study offers empirical insights into the advantages and limitations of different modeling approaches and provides recommendations for selecting appropriate methods based on dataset size, interpretability needs, and computational constraints. © 2025, CC BY.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Speech Separation for Low-Resource Languages

Speech Separation for Low-Resource Languages

引用

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

作者： Marvin Borsdorf Zexu Pan Pascal Himmelmann Haizhou Li Tanja Schultz Machine Listening Lab University of Bremen Bremen Germany Alibaba Group Singapore Department of Mathematics and Computer Science FernUniversität in Hagen Hagen Germany School of Data Science Shenzhen Research Institute of Big Data The Chinese University of Hong Kong Shenzhen China Department of Electrical and Computer Engineering National University of Singapore Singapore Cognitive Systems Lab University of Bremen Bremen Germany

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Speech separation aims to equip machines with the human ability of selective listening, i.e. to focus attention on specific information in spoken communication. Studies have shown that the language spoken in a cocktail party scenario matters. While the development of speech separation models can leverage extensive databases, for the majority of languages only very limited data is available. This work presents the very first study on speech separation for low-resource languages. We choose blind source separation as the task to be studied and analyze three strategies to overcome the data scarcity of two low-resource languages from the GlobalPhoneMS2 database. We show that data from other languages can be used to develop models that work for low-resource languages. Finetuning additionally boosts the performance, and training on multiple languages increases both performance and robustness. We show that dynamic mixing in the development helps to find a trade-off between performance and development time.

关键词： Training Measurement databases Signal processing data models Blind source separation Robustness Acoustics Speech processing

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：