检索结果-内蒙古大学图书馆

您好，读者！请登录

内蒙古大学图书馆

首页
概况
党建
资源
服务
科研支持
- 论文收录引用证明
- 科技查新
知识产权
档案馆
帮助

咨询与建议

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

您的常用邮箱：*

您的手机号码：*

问题描述：

当前已输入0个字，您还可以输入200个字

全部搜索
期刊论文
图书
学位论文
标准
纸本馆藏
外文资源发现
数据库导航
超星发现

高级检索

时间限定

出版年份：

文献类型

图书期刊文献学位论文多媒体

馆藏选择

电子馆藏纸本馆藏

核心期刊

全部期刊 SCI 收录期刊 SSCI 收录期刊 EI 收录期刊 CSCD 收录期刊 CSSCI 收录期刊

语言

中文英文

文献类型

期刊文献图书学位论文标准纸本馆藏

帮助

文字说明：

T=题名（书名、题名），A=作者（责任者），K=主题词，P=出版物名称，PU=出版社名称，O=机构（作者单位、学位授予单位、专利申请人），L=中图分类号，C=学科分类号，U=全部字段，Y=年（出版发行年、学位年度、标准发布年）

检索规则说明：

AND代表“并且”；OR代表“或者”；NOT代表“不包含”；(注意必须大写,运算符两边需空一格)

检索范例：

范例一：(K=图书馆学 OR K=情报学) AND A=范并思 AND Y=1982-2016
范例二：P=计算机应用与软件 AND (U=C++ OR U=Basic) NOT K=Visual AND Y=2011-2016

分类表

所选分类

>> <<

限定检索结果

文献类型

36,643 篇 会议
343 篇 期刊文献
31 册 图书
1 篇 学位论文

馆藏范围

37,018 篇 电子文献
1 种 纸本馆藏

日期分布

学科分类号

21,227 篇 工学
- 15,943 篇 计算机科学与技术...
- 8,292 篇 软件工程
- 6,694 篇 电气工程
- 4,501 篇 控制科学与工程
- 4,291 篇 信息与通信工程
- 1,606 篇 电子科学与技术（可...
- 1,420 篇 机械工程
- 1,034 篇 仪器科学与技术
- 802 篇 动力工程及工程热...
- 707 篇 交通运输工程
- 705 篇 生物医学工程（可授...
- 690 篇 生物工程
- 655 篇 安全科学与工程
- 445 篇 光学工程
- 377 篇 网络空间安全
- 329 篇 航空宇航科学与技...
- 326 篇 建筑学
- 308 篇 石油与天然气工程
4,277 篇 理学
- 2,098 篇 数学
- 1,020 篇 物理学
- 812 篇 系统科学
- 747 篇 生物学
- 517 篇 统计学（可授理学、...
3,409 篇 管理学
- 2,828 篇 管理科学与工程(可...
- 1,223 篇 工商管理
- 811 篇 图书情报与档案管...
794 篇 医学
- 645 篇 临床医学
- 439 篇 基础医学(可授医学...
338 篇 法学
- 301 篇 社会学
323 篇 经济学
- 321 篇 应用经济学
163 篇 农学
136 篇 教育学
79 篇 军事学
62 篇 文学
32 篇 艺术学

主题

5,962 篇 real-time system...
4,007 篇 real time system...
2,278 篇 embedded systems
1,843 篇 hardware
1,383 篇 computer archite...
1,360 篇 computational mo...
1,282 篇 control systems
1,239 篇 application soft...
1,103 篇 embedded system
1,077 篇 monitoring
825 篇 protocols
805 篇 accuracy
703 篇 cameras
666 篇 processor schedu...
631 篇 cloud computing
621 篇 wireless sensor ...
614 篇 robot sensing sy...
606 篇 field programmab...
594 篇 internet of thin...
585 篇 training

机构

57 篇 ieee
34 篇 school of comput...
29 篇 malardalen univ ...
23 篇 zhejiang univ co...
22 篇 inria
22 篇 institute of inf...
22 篇 tsinghua univers...
22 篇 centre for high ...
21 篇 mälardalen real-...
21 篇 lovely professio...
20 篇 department of co...
20 篇 department of el...
19 篇 college of compu...
19 篇 beijing key labo...
19 篇 natl univ singap...
19 篇 eindhoven univer...
18 篇 univ chinese aca...
18 篇 institute of inf...
17 篇 beijing key labo...
16 篇 nanyang technol ...

作者

49 篇 obermaisser roma...
24 篇 moris behnam
24 篇 rinner bernhard
21 篇 liu yang
19 篇 jian-jia chen
17 篇 catthoor francky
17 篇 bril reinder j.
17 篇 alois knoll
16 篇 hao zhang
15 篇 mohamed abid
15 篇 takada hiroaki
15 篇 li li
15 篇 shinpei kato
14 篇 lei wang
14 篇 rui wang
14 篇 dakai zhu
14 篇 rajkumar buyya
14 篇 dustdar schahram
14 篇 yi wang
14 篇 thomas nolte

语言

35,103 篇 英文
1,751 篇 其他
147 篇 中文
8 篇 土耳其文
6 篇 西班牙文
2 篇 法文
2 篇 葡萄牙文
2 篇 俄文
1 篇 捷克文

检索条件"任意字段=IEEE International Conference on Embedded and Real-Time Computing Systems and Applications"

共 37018 条记录，以下是1611-1620 订阅

全选清除本页清除全部题录导出标记到"检索档案"

详细简洁

排序：

RTiL: real-time Inference of Large Language Models on Memory-Constrained GPU Devices

RTiL: Real-Time Inference of Large Language Models on Memory...

引用

ieee international conference on embedded and real-time computing systems and applications (RTCSA)

作者： Juxin Niu Wei Zhang Chun Jason Xue Nan Guan City University of Hong Kong Hong Kong SAR Shandong University China Mohamed bin Zayed University of Artificial Intelligence UAE

ISBN: (数字)9798350387957

ISBN: (纸本)9798350387964

While large language models (LLMs) are usually deployed on powerful servers, there is growing interest in deploying them on local machines for better real-time performance, service stability, privacy, and flexibility. Unfortunately, the GPU memory on local machines is often insufficient to accommodate the entire LLM. Although running an LLM on such a GPU device is still possible by swapping data between the limited GPU memory and the abundant main memory, the slow speed of data swapping significantly hampers inference time, rendering it impractical in reality. In this paper, we propose RTiL, a systematic solution to address the above challenge. RTiL utilizes collaborative inference, which combines a lightweight LLM with the default powerful LLM. The lightweight LLM generates output tokens, which are then validated for quality by the powerful LLM. This approach allows RTiL to significantly speed up inference while maintaining the same output quality as when using the powerful LLM alone. Additionally, by delegating part of the inference workload to the CPU and optimizing data movement between main and GPU memory, we further enhance the efficiency of the inference process. Furthermore, we extend RTiL to handle requests with real-time requirements, enabling it to meet such demands by slightly trading off output quality. Through extensive experiments, we demonstrate notable improvements in inference efficiency and the ability to fulfill real-time requirements while minimizing degradation in output quality.

关键词： Performance evaluation Privacy Systematics Large language models Memory management Graphics processing units Rendering (computer graphics) real-time systems Stability analysis Servers

来源：评论

学校读者我要写书评

暂无评论

Edge-based Joint User Association and Resource Allocation for Light Field Metaverse systems

Edge-based Joint User Association and Resource Allocation fo...

引用

2023 ieee international conference on Metaverse computing, Networking and applications, MetaCom 2023

作者： Wang, Chenchen Hu, Xinjue Zhang, Lin Beijing University of Posts and Telecommunications School of Artificial Intelligence Beijing China Wuhan University of Technology Intelligent Transportation System Research Center Wuhan China Beijing Big Data Center Beijing China

ISBN: (纸本)9798350333336

Light field (LF), as the representative of Metaverse multimedia format, its huge data scale and heavy rendering burden have brought great challenges to networks and user devices. Although edge-based light field Metaverse systems can alleviate the enormous pressure of computing on users' devices, it will also make users need to pay additional charges of edge computing power and mobile data traffic for low-latency, high-quality light field Metaverse services. In this poster, we study the user association and bandwidth/computational resource allocation for multiple light field users connected to multiple edge nodes environment. An online Joint User Association and Resources Allocation (JUARA) optimization algorithm based on deep reinforcement learning is proposed, which can balance the service fee cost by users and response delay of LF service in a long-term time-varying network environment. The viewpoint rendering delay, data traffic fees and edge computing rental fees are evaluated in the simulation, and the results demonstrate the effectiveness of edge-assisted LF transmission architecture and proposed JUARA scheduling algorithm. © 2023 ieee.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Pedestrian Detection: An Explainable Approach 23

Pedestrian Detection: An Explainable Approach

引用

23rd ieee international conference on Machine Learning and applications, ICMLA 2024

作者： Pang, Hongbo Huang, Changcheng Carleton University Department of Systems and Computer Engineering Ottawa Canada

ISBN: (纸本)9798350374889

In the rapidly advancing field of autonomous technologies, particularly in transportation and surveillance, there is a critical need for accurate and reliable pedestrian detection systems. While deep learning models have achieved remarkable success in this area, their 'black box' nature often concerns their trustworthiness and practical applicability. Our study introduces a modified YOLOv5s model, augmented with a Global Attention Mechanism and enhanced with Gradient-weighted Class Activation Mapping (Grad-CAM) for improved explainability. This approach maintains the high accuracy typical of deep learning models while significantly increasing their transparency. Our modified YOLOv5s architecture is tailored for real-time applications, achieving an inference speed of 9.4 milliseconds on a single NVIDIA A100 GPU. Incorporating Grad-CAM provides a more intuitive interpretation of the model's decisions, potentially increasing user confidence in its predictions. We evaluate our enhanced framework on the CityPersons dataset, demonstrating improvements in precision, recall, and mAP@0.5 compared to the baseline YOLOv5s model. Our results indicate a promising advancement in pedestrian detection, balancing high performance with improved explainability. © 2024 ieee.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

international conference on VLSI, Signal Processing, Power Electronics, IoT, Communication and embedded systems, VSPICE 2023

International Conference on VLSI, Signal Processing, Power E...

引用

international conference on VLSI, Signal Processing, Power Electronics, IoT, Communication and embedded systems, VSPICE 2023

ISBN: (纸本)9789819746569

The proceedings contain 35 papers. The special focus in this conference is on VLSI, Signal Processing, Power Electronics, IoT, Communication and embedded systems. The topics include: Sensor Perception Reliability in Automated Vehicles;design and Implementation of 16-Bit Posit Arithmetic;emergence, Evolution, and applications of Medical Cyber-Physical systems;smart Insole Design for Foot Pressure Monitoring;an Efficient Cryptographic Technique to Improve Data Privacy and Security in E-Commerce;an Iot-Based Transmission Line Fault Monitoring System;performance Analysis of the System of IoT Architecture;real-time Implementation of IoT-Based Home Automation System on Arduino Platform;Data Security Paradigms: Paillier Homomorphic Encryption and AES in the Context of Privacy-Preserving Computation;a Review on Watermarking Techniques of Digital Image;An Improved Pre-charging Strategy for Asymmetrical Mixed-Cell Submodule (SM) Based Modular Multilevel Converter;classification of Remotely Sensed Data Using Fisher’s Linear Discriminant;design of Smart Wheelchair for Disabilities with Quadriplegia Through Brain–Computer Interface;Automatic Load Distribution of Power During Low Power Generation Using PLC;AI-driven Lung Cancer Detection for Rapid Analysis of Medical Imaging Data;comparative Study of Transfer Learning Models for Mushroom Classification;automated Software Defect Prediction Model: AdaBoost-Based Support Vector Machine Approach;Preprocessing of MRI Data for the Early Detection of Alzheimer’s Disease Using DWT;classification of Soybean Seed Using Support Vector Machine with Image Enhancement Techniques;From Deployment to Drift: A Comprehensive Approach to ML Model Monitoring with Evidently AI;a real-time Computer Vision-Based Green Intensity Estimation Method for the Arecanut Tree Pesticide Spraying System;key Enabling Technologies for Next-Generation Wireless systems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Mean-reverting Prices and Losses in Liquidity Pools 6

Mean-reverting Prices and Losses in Liquidity Pools

引用

6th international conference on Blockchain computing and applications, BCCA 2024

作者： Ladóczki, Bence HUN-REN-BME Information Systems Research Group HSNLab Budapest University of Technology and Economics Hungary

ISBN: (纸本)9798350351538

With the proliferation of decentralised finance (DeFi) protocols AMMs (automated market makers) on blockchains have also gained momentum in the past few years. This is mostly attributable to the fact that they require much weaker trust assumptions than centralised exchanges (CEXs) and that they can be utilised anonymously without requiring their users to go through a know-your-customer process. The decentralised nature of AMMs brings in some caveats, among which is the intensively researched loss-versus-rebalancing (LVR) metric, which is the dollar amount a liquidity pool (LP) loses in one unit of time compared to a rebalancing strategy that executes the same trades on CEXs. In this work, we adopt the Ornstein-Uhlenbeck stochastic differential equation to account for uncertain movements in the spot price, develop a discrete-time simulator and gauge the LVR metrics of constant function market makers (CFMMs). We demonstrate that the relative amount of losses is not as large as one would expect at first glance. © 2024 ieee.

关键词： Decentralized finance

来源：评论

学校读者我要写书评

暂无评论

Chatbots Lifecycle Support Platform 12

Chatbots Lifecycle Support Platform

引用

12th ieee international conference on Intelligent Data Acquisition and Advanced computing systems: Technology and applications, IDAACS 2023

作者： Nowakowski, Grzegorz Telenyk, Sergii Vovk, Yevhenii Cracow University of Technology Faculty of Electrical and Computer Engineering Cracow Poland National Technical University of Ukraine Igor Sikorsky Kyiv Polytechnic Institute Kyiv Ukraine

ISBN: (纸本)9798350358056

This article proposes the concept of a platform for the development, accumulation and use of specialised applications - chatbots - that automate functions related to providing information, placing orders and fulfilling orders, and the implementation of multi-stage processes using the capabilities of social networks and messaging programs. This platform is built on the basis of the approach, according to which chatbots are built in several stages. For each of the stages, there are formal models and methods that ensure a high level of automation of work on the creation, development and maintenance of chatbots. Thus, the creation of a chatbot consists in a chain of automated transformations of business processes into transactional systems that allow describing the corresponding state machines, and state machines into formal means of their description in popular messengers. Natural language processing models are used to communicate with the bot and identify user intent. The stages of describing the bot's state machine and the transition to the formal definition of a chatbot in the messenger are demonstrated on a real example. © 2023 ieee.

关键词： bots formal systems IT systems lifecycle support platform

来源：评论

学校读者我要写书评

暂无评论

IIOT and Deep Learning for Decision-Making: Heading Towards Intelligent Maintenance 10

IIOT and Deep Learning for Decision-Making: Heading Towards ...

引用

10th international conference on Optimization and applications, ICOA 2024

作者： Chaibi, Adham Stitou, Ahmed Lagrat, Ismail Serrou, Driss El Mhamedi, Abderrahman Tchoffa, David Oussama, Bouazaoui Labiba, Bousmaki National School of Applied Sciences Ibn Tofail University Advanced Systems Engineering Lab Kenitra Morocco Iut of Montreuil Paris 8 University Qartz Laboratory Paris France

ISBN: (纸本)9798350387353

Traditional maintenance (TM) practices in industrial companies reveal several drawbacks. It is predominantly conducted on fixed schedules rather than based on actual equipment conditions, resulting in issues such as downtime and additional maintenance costs. These issues stem from a lack of real-time data and precise diagnostics, making early detection of failures challenging. This underscores the increasing importance of Industry 4.0 (I4), particularly Predictive maintenance (PM) or what is known as Maintenance 4.0. The latter leverages real-time operational data and sophisticated techniques like data analytics, Machine learning (ML), and peep learning (DL) to anticipate failures in advance. This approach helps reduce maintenance costs and unplanned downtime. © 2024 ieee.

关键词： Predictive maintenance

来源：评论

学校读者我要写书评

暂无评论

Deep Learning for real-time Drowsiness Onset Detection 2

Deep Learning for Real-Time Drowsiness Onset Detection

引用

2nd ieee international conference on Integrated Circuits and Communication systems, ICICACS 2024

作者： Govindaraj, M. Asha, V. Katingeri, Mohammed Mazil Manvanth, L.J. Ch, Mouli Raghavendra Prashanth Mathew, Jincy C New Horizon College of Engineering Master of Computer Applications Bengaluru India

ISBN: (纸本)9798350317558

Drowsiness detection is a pivotal element in ensuring the well-being of both drivers' and passengers safety, as it helps prevent accidents caused by tired or drowsy individuals behind the wheel. This research introduces an advanced hybrid model to effectively and swiftly identify drowsiness in real-time. The process commences with data pre-processing, involving the capture and examination of facial images of the driver for analysis. These images go through different pre-processing stages, including resizing, converting to grayscale, and reducing noise through a blend of filters, such as Weiner and Median filters. Our model similar to ResNet, is meticulously fine-tuned and optimized using sophisticated techniques, including Aquila optimization, to enhance its performance. The primary objective of this model is to ascertain whether the driver is in a drowsy or alert state. The model's performance is rigorously assessed and benchmarked against previous research in the field. The attained outcomes by our proposed drowsiness detection model are truly impressive, boasting an F1 score of 98.98%, an accuracy rate of 99.2%, a sensitivity rate of 98.9%, a specificity rate of 99.1%, and a precision rate of 99%. These exceptional metrics underscore the model's effectiveness in accurately identifying drowsiness, underscoring its importance in promoting road safety. © 2024 ieee.

关键词： Accident prevention Convolutional Neural Networks Eye tracking Onset Detection real-time Detection time cumulative effect

来源：评论

学校读者我要写书评

暂无评论

Implementing IoT and Deep Learning-Based Educational systems for Assessing Practical Learning with Cloud computing applications 4

Implementing IoT and Deep Learning-Based Educational Systems...

引用

4th international conference on Technological Advancements in Computational Sciences, ICTACS 2024

作者： Thakur, Hardeo Kumar Bafna, Minal Singh, Jagendra Naresh, Usikela Biswas, Susmi Kiran, Saggurthi School of Computer Science Engineering and Technology Bennett University Greater Noida India Agrawal Pg College Department of Physics Jaipur India Rangareddy India Haldia Institute of Management Makaut Department of Management Science West Bengal India Hyderabad India

ISBN: (纸本)9798350387490

In this study, the practical application of technologies related to IoT and machine learning in the robotics education context is examined based on the use of a robot with two proximity sensors. The trial system was designed to leverage the real-time sensor data and monitor and optimize student interactions with the robot in the course of performing such tasks as object manipulation and palletization. The sensor data is used to critically track the accuracy of student actions and help mentors provide feedback on the issue. The machine learning models used in the study, which are Recurrent Neural Network (RNN), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Long Short-Term Memory (LSTM), are applied to analyze the sensor data and predict and classify the student's actions so that corrective measures can be taken in time. The RNN model has the highest accuracy and efficiency for the task, as it reaches the level of 96.7% compared to 80% for KNN, 71% for SVM, and 75% for LSTM. This advantage is explained by the capability to model temporal dependencies or otherwise capture with its memory the sequences present in the sensor data. The data is transmitted securely and efficiently to the cloud platforms and back due to the implementation of the IoT, which also helps to deploy the model more effectively. Overall, it is concluded that the integration of IoT and machine learning is beneficial and transformative in educational contexts as it provides students with personalized real-time feedback. In this way, the level of proficiency of the students in dealing with robotic and IoT-based systems is significantly enhanced, preparing them to successfully address the future challenges of industry and academia in the sphere. © 2024 ieee.

关键词： Proximity sensors

来源：评论

学校读者我要写书评

暂无评论

A Reservation-Based List Scheduling for embedded systems with Memory Constraints 23rd

A Reservation-Based List Scheduling for Embedded Systems wi...

引用

23rd international conference on Parallel and Distributed computing, applications, and Technologies, PDCAT 2022

作者： Wang, Kai-Siang Chou, Jerry National Tsing Hua University Hsinchu Taiwan

ISBN: (纸本)9783031299261

Many embedded systems have hard resource constraints that make schedules found by list scheduling heuristics infeasible. One of the main challenges yielded by memory constraints and the high degree of parallelism is deadlock. In this paper, our primary goal is to find a feasible solution given the memory constraints. We propose a reservation-based solution, an extension for list scheduling algorithms, that can be integrated into those algorithms and make them aware of memory constraints. We show our technique prevents deadlock and significantly reduces the required memory size. The experimental results on randomly generated graphs and real world applications show that our proposed solution can obtain relatively high-quality solutions with up to 10% makespan improvement and 30% memory reduction on average. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： embedded systems

来源：评论

学校读者我要写书评

暂无评论

没有更多数据了...

全选清除本页清除全部题录导出标记到“检索档案”

共500页 << < 158 159 160 161 162 163 164 165 166 167 > >>

检索报告对象比较合并检索0

隐藏清空

合并搜索

回到顶部

执行限定条件

内容：

评分：

请选择保存的检索档案：

请选择收藏分类：

订阅名称：

通借通还

温馨提示：

图书名称：

借书校区：

取书校区：

手机号码：

邮箱地址：

一卡通帐号：

电话和邮箱必须正确填写，我们会与您联系确认。

联系人：

所在院系：

联系邮箱：

联系电话：

内蒙古自治区呼和浩特市赛罕区大学西街235号邮编: 010021

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：