检索结果-内蒙古大学图书馆

10th IEEE Smart World Congress, SWC 2024

作者： Guo, Jinpeng Fan, Feiyi Liu, Renju Jin, Yuan Zhang, Wei Chen, Yiqiang Gu, Yang Zhengzhou University Henan Institute of Advanced Technology Zhengzhou China Chinese Academy of Sciences Institute of Computing Technology Beijing China The Beijing Key Laboratory of Mobile Computing and Pervasive Device Beijing China Intelligent Technology Co. Ltd. Shenzhen China

ISBN: (纸本)9798331520861

Cuff-less blood pressure estimation methods usually require personalized fine-tuning to achieve better estimation accuracy due to significant inter-individual data variability. The emerging domain generalization methods can extract blood pressure-related knowledge from diverse individuals and estimate readings for unseen subjects;however, these existing methods are less effective in aligning distributions for regression tasks on highly individual physiological time-series data. In this paper, we present a meta-learning-based approach for domainin-variant blood pressure estimation (MetaDIBP), which optimizes a temporal invariant representation learning module with a twostage meta-learning strategy. Specifically, we minimize differences between a prospective target domain and the rest of the source domains in the meta-training stage;update the model utilizing all source domains to ensure consistent and robust performance across various unseen target domains in the successive metatesting stage;and incorporate deep feature extraction, Correlation Alignment loss, and Maximum Mean Discrepancy loss into a temporal invariant representation learning module to serve as the network architecture for feature distribution discrepancy reduction. Our experimental results demonstrate that the proposed method outperforms traditional domain generalization and deep-learning-based blood pressure models, achieving a mean absolute error of 12.989 ~ m m H g for systolic blood pressure and 7.391 mmHg for diastolic blood pressure. With personalized finetuning, the mean absolute error further improves to 7.464 mmHg for systolic blood pressure and 4.509 ~ m m H g for diastolic blood pressure, showing advantages over existing methods. © 2024 IEEE.

关键词： Blood pressure

来源：评论

学校读者我要写书评

暂无评论

Improving Performance of Virtual Machine Covert Timing Channel Through Optimized Run-Length Encoding

引用

Journal of Computer Science & technology 2023年第4期38卷 793-806页

作者：王翀陈荣亮古亮 Shenzhen Institute of Advanced Technology Chinese Academy of SciencesShenzhen 518055China Sangfor Technologies Inc. Shenzhen 518055China Shenzhen Key Laboratory for Exascale Engineering and Scientific Computing Shenzhen 518055China

With its wider acceptability,cloud can host a diverse set of data and applications ranging from entertainment to personal to *** foundation of cloud computing is based on virtual machines where boundaries among the application data are very thin,and the potential of data leakage exists all the *** instance,a virtual machine covert timing channel is an aggressive mechanism to leak confidential information through shared components or networks by violating isolation and security policies in *** performance of a covert timing channel(covert channel)is crucial to adversaries and attempts have been made to improve the performance of covert timing channels by advancing the encoding mechanism and covert information *** promising,the redundancy of the covert message is mainly *** paper applies three encoding schemes namely run-length,Huffman,and arithmetic encoding schemes for data compression of a virtual machine covert timing channel by exploiting ***,the paper studies the performance of such channels according to their ***,we show that these encoding schemes still contain redundancy in a covert channel scenario,and thereby a new encoding scheme namely optimized Runlength encoding(OptRLE)is presented that greatly enhances the performance of a covert timing *** optimizations schemes adopted by OptRLE are also discussed,and a mathematical model of the behavior of an OptRLE-based covert timing channel is *** theoretical capacity of a channel can be obtained using the proposed *** analysis reveals that OptRLE further improves the performance of a covert timing channel,in addition to the effects of the *** result shows how OptRLE affects the size of covert data and the capacity of covert timing channels,and why the performance of the covert timing channel is improved.

关键词： covert storage channel information security covert channel threat evaluation anti-detection criterion covert channel restriction

来源：评论

学校读者我要写书评

暂无评论

A Data-driven Spatiotemporal Simulator for Reinforcement Learning Methods 50th

A Data-driven Spatiotemporal Simulator for Reinforcement Lea...

引用

50th International Conference on Very Large Data Bases, VLDB 2024

作者： Shi, Dingyuan Song, Bingchen Zhang, Yuanyuan Yang, Haolong Xu, Ke Beijing Advanced Innovation Center for Future Blockchain and Privacy Computing Beihang University China Beijing Institute of Astronautical Systems Engineering China North China Institute of Computing Technology China Beihang University China

Spatiotemporal applications such as taxi order dispatching and warehouse task scheduling depend critically on the algorithms for operational efficiency. However, the inherent dynamic nature of these applications presents challenges in algorithm design. The growth of mobility services has facilitated the collection of extensive spatiotemporal data, which in turn prompted algorithm designers to use data-driven methods. Reinforcement learning (RL), recognized for its strong performance and suitability for spatiotemporal contexts, has garnered considerable research interest. Despite their potential, RL algorithms necessitate the use of a simulator for both training and validation purposes. However, no specific simulation system has been developed for spatiotemporal algorithm design. This vacancy hinders the progress of spatiotemporal algorithm designers. In this demo, we build a system called Data-driven Spatiotemporal Simulator (DSS), hoping to bring convenience for spa-tiotemporal algorithm designers. DSS is adept at handling problems related to taxi order dispatching and warehouse task scheduling and possesses the versatility to be expanded for other user-defined scenarios. The system includes visualization modules that offer insightful panels, alongside developer tools designed to streamline the development process. This enables designers to efficiently craft, evaluate, and refine their algorithms, potentially accelerating innovation in spatiotemporal application development. © 2024, VLDB Endowment. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

advanced 3D-printed antibacterial nano 58S bioglass/AgNPs/CeO2-based scaffolds for bone tissue engineering

引用

Ceramics International 2025年第15期51卷 20939-20953页

作者： Singh, Indrajeet Kong, Ing Viswakarma, Ashok Balani, Kantesh Advanced Polymer and Composite Materials Laboratory Department of Engineering School of Computing Engineering and Mathematical Sciences La Trobe University BendigoVIC3550 Australia Department of Materials Science and Engineering Indian Institute of Technology Kanpur208016 India Department of Mechanical Engineering Indian Institute of Technology Kanpur208016 India

58S bioglass (BG) has been considered an effective bioceramic in bone tissue engineering due to its good osteoinductive and osteoconductive properties;however, its poor antibacterial and printing properties remain a challenge in the development of ideal patient-specific, fully interconnected, bone-like scaffolds. To address this issue, this study presents novel 3D-printed BG/AgNPs/CeO2 (BGAC3D) composite scaffolds. These scaffolds exhibit a fully interconnected pore network (≈99 %) with a porosity of ≈64 % and optimally designed pore and strut sizes of 404 ± 15 μm and 394 ± 30 μm, respectively, making them suitable for bone healing applications. Adding AgNPs and ceria nanoparticles significantly increased the shear-thinning properties of Pluronic F-127/BG/AgNPs/CeO2-based ink, as compared to Pluronic F-127/BG-based ink. As a result, BGAC3D scaffold showed improved dimensional accuracy during printing relative to pristine BG. Nanoindentation studies revealed substantial increase in the mechanical properties of BGAC3D scaffolds, with elastic modulus, hardness, and compressive strength showing enhancement of 76 % (57.9 ± 8 GPa), 24 % (3.1 ± 1 GPa), and 275 % (1.5 ± 0.4 MPa), respectively. The antibacterial properties of BGAC3D scaffolds were investigated using disc diffusion and bacterial adhesion assays, which demonstrated an exceptional bactericidal efficacy of 99.3 % against Escherichia coli (E. coli) and 99.4 % against Staphylococcus aureus (S. aureus) compared to pristine BG. The cytocompatibility of the scaffolds was assessed using resazurin assays with mouse BMSCs and MTT assays with mouse fibroblasts, revealing enhanced cell viability for the BGAC3D composition compared to BG. Additionally, both BG and BGAC3D composites showed similar apatite mineralization on their surfaces, as observed under a scanning electron microscope. These findings indicate that BGAC3D scaffolds possess the necessary mechanical, antibacterial, and cytocompatibility properties, making th

关键词： Escherichia coli

来源：评论

学校读者我要写书评

暂无评论

Large Language Models Can Automatically Engineer Features for Few-Shot Tabular Learning 41

Large Language Models Can Automatically Engineer Features fo...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Han, Sungwon Yoon, Jinsung Arik, Sercan Ö. Pfister, Tomas School of Computing Korea Advanced Institute of Science and Technology Daejeon Korea Republic of Google Cloud AI SunnyvaleCA United States

Large Language Models (LLMs), with their remarkable ability to tackle challenging and unseen reasoning problems, hold immense potential for tabular learning, that is vital for many real-world applications. In this paper, we propose a novel in-context learning framework, FeatLLM, which employs LLMs as feature engineers to produce an input data set that is optimally suited for tabular predictions. The generated features are used to infer class likelihood with a simple downstream machine learning model, such as linear regression and yields high performance few-shot learning. The proposed FeatLLM framework only uses this simple predictive model with the discovered features at inference time. Compared to existing LLM-based approaches, FeatLLM eliminates the need to send queries to the LLM for each sample at inference time. Moreover, it merely requires API-level access to LLMs, and overcomes prompt size limitations. As demonstrated across numerous tabular datasets from a wide range of domains, FeatLLM generates high-quality rules, significantly (10% on average) outperforming alternatives such as TabLLM and STUNT. Copyright 2024 by the author(s)

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Based Algorithmic Trading Based on News and Events Strategies 7th

Deep Learning-Based Algorithmic Trading Based on News and Ev...

引用

7th International Conference on Soft computing: Theories and Applications, SoCTA 2023

作者： Kayathri, V. Prabakaran, P. Department of Computer Applications School of Computing Sciences Vels Institute of Science Technology and Advanced Studies Pallavaram Chennai600 117 India

ISBN: (纸本)9789819720309

Successful stock market investors should consider these trends. This article defines price effect, optimal execution, and placement through the lens of algorithmic trading and strategies based on news and events. By using feedbacks controlled to include consideration of a stop-loss order, algorithmic trading allows traders and investors at large to liquidate or construct big securities positions in a fully automated manner, thus generalising the existing stock trading results. Trading decisions made by the AI agent could affect the asset price. We see this as a marketplace where people are bartering and swapping goods. Creating a new market is the driving force behind our research into generative models. This news sentiment does a good job of describing stocks, and AI systems that read the news and quantify them react to factors that affect prices far more quickly than humans do. All three of the stock market's underpinnings—price, volume, and volatility—respond strongly to news events. Sentiment analysis is a method for gauging the overall tone of a news piece and classifying it as optimistic, pessimistic, or neutral. The simulation is conducted in python to test the effectiveness of the model in trading in an automated manner for stocks. The results show that the proposed sentimental analysis for stock trading achieves higher rate of accuracy and precision than the other methods. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Financial markets

来源：评论

学校读者我要写书评

暂无评论

Deep neural network-enabled battery open-circuit voltage estimation based on partial charging data

引用

Journal of Energy Chemistry 2024年第3期90卷 120-132,I0005页

作者： Ziyou Zhou Yonggang Liu Chengming Zhang Weixiang Shen Rui Xiong State Key Laboratory of Mechanical Transmission for Advanced Equipment&College of Mechanical and Vehicle Engineering Chongqing UniversityChongqing 400000China Department of Vehicle Engineering School of Mechanical EngineeringBeijing Institute of TechnologyBeijing 100081China Department of Electrical Engineering Harbin Institute of TechnologyHarbin 150000HeilongjiangChina School of Science Computing and Engineering TechnologiesSwinburne University of TechnologyHawthornVictoria 3122Australia

Battery management systems(BMSs) play a vital role in ensuring efficient and reliable operations of lithium-ion *** main function of the BMSs is to estimate battery states and diagnose battery health using battery open-circuit voltage(OCV).However,acquiring the complete OCV data online can be a challenging endeavor due to the time-consuming measurement process or the need for specific operating conditions required by OCV estimation *** addressing these concerns,this study introduces a deep neural network-combined framework for accurate and robust OCV estimation,utilizing partial daily charging *** incorporate a generative deep learning model to extract aging-related features from data and generate high-fidelity OCV *** analysis is employed to identify the optimal partial charging data,optimizing the OCV estimation precision while preserving exceptional *** validation results,using data from nickel-cobalt-magnesium(NCM) batteries,illustrate the accurate estimation of the complete OCV-capacity curve,with an average root mean square errors(RMSE) of less than 3 *** this level of precision for OCV estimation requires only around 50 s collection of partial charging *** validations on diverse battery types operating under various conditions confirm the effectiveness of our proposed *** cases of precise health diagnosis based on OCV highlight the significance of conducting online OCV *** method provides a flexible approach to achieve complete OCV estimation and holds promise for generalization to other tasks in BMSs.

关键词： Lithium-ion battery Open-circuit voltage Health diagnosis Deep learning

来源：评论

学校读者我要写书评

暂无评论

Mukhtasir-Khail-Net: An Ultra-Efficient Convolutional Neural Network for Sports Activity Recognition with Wearable Inertial Sensors 4

Mukhtasir-Khail-Net: An Ultra-Efficient Convolutional Neural...

引用

4th International Conference on Digital Futures and Transformative Technologies, ICoDT2 2024

作者： Imran, Hamza Ali Muhammad, Shaida Wazir, Saad Ikram, Ataul Aziz Arshad, Obaidullah National University of Sciences & Technology Department of Computing Islamabad Pakistan Korea Advanced Institute of Science & Technology School of Computing Daejeon Korea Republic of National University of Computer & Emerging Sciences School of Engineering Islamabad Pakistan

ISBN: (纸本)9798350391862

The current prevalent approach of the Internet of Health and Medical Things entails proactively preventing disease onset through routine monitoring of individuals' physical activities, making Human Activity Recognition (HAR) and Behaviour Analysis an important field of study. Sports Activity Recognition (SAcR) is a subset of HAR that focuses on identifying athletic movements. Computer Vision, Environmental Sensors, and Wearable Sensors are the three main methods for monitoring such activities. When the merits and limitations of these methods are considered, Wearable Sensors emerge as the most practical option. This study introduces the Mukhtasir-Khail-Net, a model with only 651 parameters that efficiently uses addition layers. When applied to six sports activities, the model achieves an impressive average accuracy of 98.865% using inertial sensor data generated by wearable sensors. For benchmark purposes, the presented model is also evaluated on the WISDM 11 dataset which is a Human Activity Recognition dataset from Fordham University and has achieved an accuracy of 94.24%. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Natural language is not enough: Benchmarking multi-modal generative AI for Verilog generation 24

Natural language is not enough: Benchmarking multi-modal gen...

引用

43rd International Conference on Computer-Aided Design, ICCAD 2024

作者： Chang, Kaiyan Chen, Zhirong Zhou, Yunhao Zhu, Wenlong Wang, Kun Xu, Haobo Li, Cangyuan Wang, Mengdi Liang, Shengwen Li, Huawei Han, Yinhe Wang, Ying SKLP Institute of Computing Technology Chinese Academy of Sciences Beijing China CICS Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences China Hangzhou Institute for Advanced Study University of Chinese Academy of Sciences China Shanghai Jiao Tong University China

ISBN: (纸本)9798400710773

Natural language interfaces have exhibited considerable potential in the automation of Verilog generation derived from high-level specifications through the utilization of large language models, garnering significant attention. Nevertheless, this paper elucidates that visual representations contribute essential contextual information critical to design intent for hardware architectures possessing spatial complexity, potentially surpassing the efficacy of natural-language-only inputs. Expanding upon this premise, our paper introduces an open-source benchmark1 for multi-modal generative models tailored for Verilog synthesis from visual-linguistic inputs, addressing both singular and complex modules. Additionally, we introduce an open-source visual and natural language Verilog query language framework to facilitate efficient and user-friendly multi-modal queries. To evaluate the performance of the proposed multi-modal hardware generative AI in Verilog generation tasks, we compare it with a popular method that relies solely on natural language. Our results demonstrate a significant accuracy improvement in the multi-modal generated Verilog compared to queries based solely on natural language. We hope to reveal a new approach to hardware design in the large-hardware-design-model era, thereby fostering a more diversified and productive approach to hardware design. © 2024 Copyright is held by the owner/author(s).

关键词： Open systems

来源：评论

学校读者我要写书评

暂无评论

Multi-Scale Time Series Segmentation Network Based on Eddy Current Testing for Detecting Surface Metal Defects

引用

IEEE/CAA Journal of Automatica Sinica 2025年第3期12卷 528-538页

作者： Xiaorui Li Xiaojuan Ban Haoran Qiao Zhaolin Yuan Hong-Ning Dai Chao Yao Yu Guo Mohammad S.Obaidat George Q.Huang the School of Intelligence Science and Technology University of Science and Technology Beijing the Beijing Advanced Innovation Center for Materials Genome Engineering the Key Laboratory of Intelligent Bionic Unmanned Systems and the Institute of Materials Intelligent Technology Liaoning Academy of Materials IEEE the Department of Computer Science Hong Kong Baptist University the School of Computer and Communication Engineering Key Laboratory of Advanced Materials and Devices for Post-Moore Chips Ministry of Education University of Science and Technology Beijing the Beijing Advanced Innovation Center for Materials Genome Engineering University of Science and Technology Beijing the School of Computer and Communication Engineering University of Science and Technology Beijing the King Abdullah Ⅱ School of Information Technology The University of Jordan the Department of Computational Intelligence the School of Computing SRM University the School of Engineering The Amity University The Hong Kong Polytechnic University

In high-risk industrial environments like nuclear power plants, precise defect identification and localization are essential for maintaining production stability and safety. However, the complexity of such a harsh environment leads to significant variations in the shape and size of the defects. To address this challenge, we propose the multivariate time series segmentation network(MSSN), which adopts a multiscale convolutional network with multi-stage and depth-separable convolutions for efficient feature extraction through variable-length templates. To tackle the classification difficulty caused by structural signal variance, MSSN employs logarithmic normalization to adjust instance distributions. Furthermore, it integrates classification with smoothing loss functions to accurately identify defect segments amid similar structural and defect signal subsequences. Our algorithm evaluated on both the Mackey-Glass dataset and industrial dataset achieves over 95% localization and demonstrates the capture capability on the synthetic dataset. In a nuclear plant's heat transfer tube dataset, it captures 90% of defect instances with75% middle localization F1 score.

关键词： Eddy current testing nondestructive testing semantic segmentation time series analysis

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：