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

13,377 篇 会议
17 篇 期刊文献
16 册 图书

馆藏范围

13,410 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

7,105 篇 工学
- 6,398 篇 计算机科学与技术...
- 5,095 篇 软件工程
- 1,730 篇 信息与通信工程
- 1,008 篇 控制科学与工程
- 735 篇 生物工程
- 721 篇 电气工程
- 667 篇 生物医学工程（可授...
- 582 篇 网络空间安全
- 511 篇 光学工程
- 451 篇 安全科学与工程
- 398 篇 电子科学与技术（可...
- 344 篇 交通运输工程
- 328 篇 动力工程及工程热...
- 288 篇 机械工程
- 267 篇 仪器科学与技术
2,667 篇 理学
- 1,298 篇 数学
- 879 篇 生物学
- 637 篇 统计学（可授理学、...
- 535 篇 物理学
- 328 篇 系统科学
- 242 篇 化学
1,877 篇 管理学
- 1,300 篇 管理科学与工程(可...
- 789 篇 图书情报与档案管...
- 573 篇 工商管理
947 篇 医学
- 687 篇 临床医学
- 681 篇 基础医学(可授医学...
- 659 篇 公共卫生与预防医...
- 264 篇 药学(可授医学、理...
508 篇 法学
- 471 篇 社会学
262 篇 经济学
- 262 篇 应用经济学
223 篇 农学
210 篇 教育学
198 篇 文学
9 篇 军事学
1 篇 艺术学

主题

2,085 篇 accuracy
1,690 篇 deep learning
1,200 篇 machine learning
1,188 篇 computational mo...
1,044 篇 training
1,026 篇 data models
968 篇 predictive model...
724 篇 real-time system...
687 篇 convolutional ne...
684 篇 feature extracti...
676 篇 machine learning...
585 篇 support vector m...
531 篇 federated learni...
484 篇 adversarial mach...
330 篇 contrastive lear...
311 篇 medical services
295 篇 random forests
289 篇 analytical model...
278 篇 prediction algor...
271 篇 neural networks

机构

105 篇 lovely professio...
101 篇 chitkara univers...
57 篇 department of co...
47 篇 school of comput...
45 篇 school of comput...
45 篇 department of co...
42 篇 the islamic univ...
42 篇 department of co...
40 篇 shandong provinc...
36 篇 college of compu...
32 篇 university of ch...
28 篇 department of co...
27 篇 department of co...
26 篇 computer science...
26 篇 graphic era deem...
25 篇 department of co...
25 篇 amrita vishwa vi...
25 篇 the islamic univ...
23 篇 department of co...
23 篇 college of intel...

作者

23 篇 satvik vats
18 篇 shiva mehta
18 篇 kukreja vinay
18 篇 vats satvik
17 篇 anjali t.
17 篇 rishabh sharma
17 篇 deepak upadhyay
16 篇 senthil kumar th...
15 篇 singh jagendra
15 篇 venugopalan manj...
13 篇 manjunatha
13 篇 rakesh kumar
12 篇 ramya maranan
11 篇 selvanayaki kola...
11 篇 s. kaliappan
10 篇 andersson karl
10 篇 muthukumaravel a...
10 篇 yang yang
10 篇 ayodeji olalekan...
10 篇 surendran r.

语言

12,140 篇 英文
1,269 篇 其他
306 篇 中文
1 篇 西班牙文

检索条件"任意字段=2024 International Conference on Computing, Machine Learning and Data Science, CMLDS 2024"

共 13410 条记录，以下是4981-4990 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：

Enhancing Lung Disease Identification with Multimodal data Fusion and Deep learning CNN Approach 5

Enhancing Lung Disease Identification with Multimodal Data F...

引用

5th international conference on Smart Electronics and Communication, ICOSEC 2024

作者： Grace Shalini, T. Shiny, G. Susan Saranya, R. Suresh Babu, P. Kavitha, R. Atheeswaran, Athiraja Srm Institute of Science and Technology School of Computing Department of Computational Intelligence Chengalpattu Kattankulathur India Sethu Institute of Technology Department of information technology Kariapatti India Velammal college of engineering and technology Department of Eee Madurai India Kalasalingam Academy of Research and Education Department of Computer Science and Engineering Krishnankoil India Bannari Amman Institute Of Technology Department Of Computer Science And Engineering Erode Sathyamangalam India Hyderabad India

ISBN: (纸本)9798331504403

Lung diseases, encompassing a wide spectrum from chronic conditions to acute infections, present a significant burden on global healthcare systems, necessitating precise and timely diagnostic methodologies. In this study, we propose a pioneering approach leveraging multimodal data fusion and deep learning Convolutional Neural Network (CNN) algorithms for enhanced identification of lung diseases. Our methodology integrates diverse data sources including medical images, clinical reports, and patient demographics to capture a comprehensive representation of lung pathology. We employ a state-of-the-art CNN architecture for medical image analysis, trained on a large-scale dataset encompassing various lung diseases and healthy lung tissue. Through extensive experimentation and validation, the approach achieved remarkable performance, with an accuracy of 94%, sensitivity of 92%, specificity of 93%, and an F1-score of 93% in accurately identifying and classifying lung abnormalities. Furthermore, we demonstrate the clinical utility of our method in providing valuable insights to healthcare professionals, aiding in timely diagnosis, prognosis, and treatment planning. Our findings highlight the potential of multimodal data fusion coupled with deep learning CNN algorithms in advancing the field of lung disease identification, offering a promising avenue for improving healthcare practices and patient outcomes in respiratory medicine. © 2024 IEEE.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

CSMAAFL: Client Scheduling and Model Aggregation in Asynchronous Federated learning 59

CSMAAFL: Client Scheduling and Model Aggregation in Asynchro...

引用

59th Annual IEEE international conference on Communications (IEEE ICC)

作者： Ma, Xiang Wang, Qun Sun, Haijian Hu, Rose Qingyang Qian, Yi Utah State Univ Dept Elect & Comp Engn Logan UT 84322 USA San Francisco State Univ Dept Comp Sci San Francisco CA USA Univ Georgia Sch Elect & Comp Engn Athens GA USA Univ Nebraska Lincoln Dept Elect & Comp Engn Omaha NE USA

ISBN: (纸本)9781728190549

Asynchronous federated learning aims to solve the straggler problem in an environment with heterogeneity, where certain clients may possess limited computational capacities, potentially leading to model aggregation delay. The core concept behind asynchronous federated learning is to empower the server to aggregate the model as soon as it receives an update from any client without waiting for updates from multiple clients or adhering to a predetermined waiting time, which is typical in synchronous mode. Because of the asynchronous setting, a potential concern is the emergence of a stale model issue, wherein slow clients might employ an outdated local model for their data training. Consequently, when these locally trained models are uploaded to the server, they may impede the convergence of the global training. Therefore, effective model aggregation strategies play a significant role in updating the global model. Besides, client scheduling is critical when heterogeneous clients with diversified computing capacities participate in the federated learning process. This work first investigates the impact of the convergence of asynchronous federated learning mode when adopting the aggregation coefficient in synchronous mode. Effective aggregation solutions that can achieve the same convergence result as in the synchronous mode are proposed, followed by an improved aggregation method with client scheduling. The simulation results in various cases demonstrate that the proposed algorithm converges with a similar level of accuracy as the classical synchronous federated learning algorithm but effectively accelerates the learning process, especially in its early stage.

关键词： Asynchronous federated learning client scheduling model aggregation

来源：评论

学校读者我要写书评

暂无评论

Research on Radio Propagation Model Based on Deep learning and Feature Images 4

Research on Radio Propagation Model Based on Deep Learning a...

引用

4th international conference on Electronic Information Engineering and Computer Technology, EIECT 2024

作者： Wang, Ran Cui, Jidong Lai, Fang Yao, Liang Li, Weiran Ai, Junjie Lv, Jiayin Henan Provincial Forestry and Ecological Construction and Development Center Zhengzhou China School of Electronic Engineering Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9798331528850

The radio propagation model based on deep learning can extract features from a large amount of relevant data and learn from them to adapt to the real-time computing requirements of complex environmental scenes. In this study, we construct a multi-dimensional feature dataset and develop a U-Net-based propagation model. To enhance the performance of the model, we introduce a feature classification preprocessing module before the model input to distinguish features with different physical meanings. Additionally, Atrous Spatial Pyramid Pooling is incorporated in the encoder to improve the model's ability to learn multi-scale features. Experiments are conducted on a dataset consisting of 275 cells in an urban area, using MAE, RMSE, and MAPE as evaluation metrics. The results demonstrate that our proposed model outperforms the original U-Net architecture in terms of all evaluation metrics for overall, indoor, and outdoor areas. © 2024 IEEE.

关键词： Contrastive learning

来源：评论

学校读者我要写书评

暂无评论

Intelligent Transportation System for Sustainable and Efficient Urban Mobility Traffic Flow Prediction Using Optimized Deep learning Approach 5

Intelligent Transportation System for Sustainable and Effici...

引用

5th international conference on Intelligent Communication Technologies and Virtual Mobile Networks, ICICV 2024

作者： Sreelekha, M. Janarthanan, Midhunchakkaravarthy Lincoln University College Faculty of Engineering Malaysia Lincoln University College Faculty of Computer Science and Multimedia Malaysia

ISBN: (纸本)9798350385649

Traffic blockage is described as the transport system's condition, characterized by the automobiles' low speed. It is also because of roads, climatic conditions, particular zones, temperature, the number of events on that day etc. It mainly happens in metropolitan cities. Due to the development of communication technology, most organizations have transferred their conventional methods to computing methods. The primary objective of this study is to investigate deep learning algorithms that can efficiently capture the inherent relationships present in network traffic flows, thereby facilitating precise predictions. To evaluate the precision and predictive potential of these approaches, two initial baseline models are constructed utilizing Recurrent Neural Network (RNN) models. These baseline models make predictions for target values' mean and mode, drawing from the training data. A model is considered accurate and proficient if it surpasses the basic, naive benchmarks established based on predefined performance metrics. The optimization process, relying on Red Fox (RF), is determined through a quantitative analysis employing well-established evaluation criteria commonly used in supervised deep learning, including accuracy, precision, recall, Mean Squared Error (MSE), and Root Mean Square Error (RMSE). © 2024 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Out-of-Domain Generalization in Dynamical Systems Reconstruction 41

Out-of-Domain Generalization in Dynamical Systems Reconstruc...

引用

41st international conference on machine learning, ICML 2024

作者： Göring, Niclas Hess, Florian Brenner, Manuel Monfared, Zahra Durstewitz, Daniel Department of Theoretical Neuroscience Central Institute of Mental Health Medical Faculty Mannheim Heidelberg University Mannheim Germany Faculty of Physics and Astronomy Heidelberg University Heidelberg Germany Interdisciplinary Center for Scientific Computing Heidelberg University Heidelberg Germany

In science we are interested in finding the governing equations, the dynamical rules, underlying empirical phenomena. While traditionally scientific models are derived through cycles of human insight and experimentation, recently deep learning (DL) techniques have been advanced to reconstruct dynamical systems (DS) directly from time series data. State-of-the-art dynamical systems reconstruction (DSR) methods show promise in capturing invariant and long-term properties of observed DS, but their ability to generalize to unobserved domains remains an open challenge. Yet, this is a crucial property we would expect from any viable scientific theory. In this work, we provide a formal framework that addresses generalization in DSR. We explain why and how out-of-domain (OOD) generalization (OODG) in DSR profoundly differs from OODG considered elsewhere in machine learning. We introduce mathematical notions based on topological concepts and ergodic theory to formalize the idea of learnability of a DSR model. We formally prove that black-box DL techniques, without adequate structural priors, generally will not be able to learn a generalizing DSR model. We also show this empirically, considering major classes of DSR algorithms proposed so far, and illustrate where and why they fail to generalize across the whole state space. Our study provides the first comprehensive mathematical treatment of OODG in DSR, and gives a deeper conceptual understanding of where the fundamental problems in OODG lie and how they could possibly be addressed in practice. Copyright 2024 by the author(s)

关键词： State space methods

来源：评论

学校读者我要写书评

暂无评论

HyperSack: Distributed Hyperparameter Optimization for Deep learning using Resource-Aware Scheduling on Heterogeneous GPU Systems 31

HyperSack: Distributed Hyperparameter Optimization for Deep ...

引用

31st Annual IEEE international conference on High Performance computing, data, and Analytics, HiPC 2024

作者： Alnaasan, Nawras Ramesh, Bharath Yao, Jinghan Shafi, Aamir Subramoni, Hari Panda, Dhabaleswar K. Department of Computer Science and Engineering The Ohio State University ColumbusOH United States

ISBN: (纸本)9798331509095

Hyperparameter Optimization (HPO) can unlock the full potential of Deep learning (DL) models;however, it is considered one of the most compute-intensive tasks in the DL domain due to multi-dimensional search spaces and complex neural network architectures. A common method for accelerating HPO workloads is parallelizing training jobs on multiple computing devices, such as modern GPUs in High-Performance computing (HPC) environments. Nonetheless, existing HPO parallelization strategies underutilize powerful GPU devices, like the NVIDIA A100 and H100, especially for training lightweight Deep Neural Networks. Resource-sharing mechanisms can improve GPU utilization;nonetheless, naïve adaptations for HPO workloads lead to poor performance. Therefore, we propose HyperSack—a distributed HPO framework for dynamic and resource-aware scheduling on heterogeneous GPU-based HPC systems with resource elasticity and fault tolerance. HyperSack reduces the execution time of HPO workloads by orchestrating the placement of DL training jobs on GPU devices with different computational capabilities. It supports different hardware architectures, HPO workloads, and scheduling policies. Our evaluations on vision and language models HPO workloads show up to 2.8x performance improvement in execution time on A100 GPUs, 4.0x on H100 GPUs, and 3.9x on a combination of 12 A100 and 4 H100 GPUs using HyperSack over standard HPO parallelization methods. © 2024 IEEE.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

FASALKA: Offloaded Privacy Classification for Blockchain Smart Contracts 6

FASALKA: Offloaded Privacy Classification for Blockchain Sma...

引用

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

作者： Devgun, Tannishtha Kumar, Gulshan Conti, Mauro School of Computer Science and Engineering Lovely Professional University Punjab India Department of Mathematics University of Padua Padua Italy International School of Advanced Studies University of Camerino Italy

ISBN: (纸本)9798350351538

Privacy problems in blockchain smart contracts arise from the inherent transparency of the technology. While blockchain ensures data integrity, it also exposes sensitive information to all participants. This lack of privacy can be a concern in industries requiring confidentiality, like finance or healthcare, prompting the need for privacy solutions in smart contracts. The existing smart contracts face the problems of over-utilization or under-utilization of privacy parameters due to the unavailability of online or offline privacy classification;smart contracts fall behind in solving privacy issues. In this paper, we introduce the first privacy classification framework for smart contracts. We call our framework oFfloaded privAcy-classified Smart contrAct for bLocKchAins (FASALKA). To be specific, FASALKA runs a novel privacy-ensured smart contract that includes a novel hybrid learning mechanism. This hybrid learning mechanism combines the potential of federated learning and reinforcement learning. We deploy Ethereum on Azure and run a set of experiments to measure the performance of our proposed FASALKA. We compare a general Ethereum framework with an Ethereum framework using our proposed FASALKA. We observe that the Ethereum framework shows 1.1% more latency than the general Ethereum;however, our proposed framework has a similar throughput of 21 TPS. Besides, FASALKA has the 100% accuracy of privacy classification, which is not present in general Ethereum. Thus, our proposed FASALKA is efficient and beneficial for the privacy-ensured blockchains. © 2024 IEEE.

关键词： Smart contract

来源：评论

学校读者我要写书评

暂无评论

Faster Recurrent Convolutional Neural Network with Edge computing Based Malware Detection in Industrial Internet of Things 2

Faster Recurrent Convolutional Neural Network with Edge Comp...

引用

2nd IEEE international conference on data science and Network Security, ICDSNS 2024

作者： Hameed Shnain, Ammar Gattupalli, Kalyan Nalini, Chekuri Alagarsundaram, Poovendran Patil, Rajeshwari College of Technical Engineering The Islamic University Department of Computers Techniques Engineering Najaf Iraq Yash Tek Inc ON Canada Mohan Babu University Erstwhile Sree Vidyanikethan Engineering College Department of Ece Tirupati India Humetis Technologies Inc KingstonNJ United States Nitte Meenakshi Institute of Technology Department of Ai & Ds Bengaluru India

ISBN: (纸本)9798350373110

The smart factory environment has become Industrial Internet of Things (1IoT) environment because of integration and openness of the environment. It is always easier to launch epidemic cyber threats in smart factories through malware most of the time, although the differentiation of exact real-time malware in these areas is complex. In order to address this issue, this research work introduces an advanced Faster Recurrent Convolutional Neural Network (Faster R-CNN) integrated with an edge computing based malware identification model. This system successfully discerns the multiple types of malwares by transmitting the huge IIoT traffic information about smart factories to the edge servers. The proposed detection system of malware consists of various layers as the edge device layer, the edge layer, lastly, the cloud layer. For edge-based deep learning it employs four base operations. The model obtained accuracy of 93.77%, recall of 95.87%, precision of 86.66%, and f1-score of 91.03% when compared to CNN and Long Short-Term Memory (LSTM). © 2024 IEEE.

关键词： Smart manufacturing

来源：评论

学校读者我要写书评

暂无评论

VeriFace: Deepfake Detector Using Deep learning 1st

VeriFace: Deepfake Detector Using Deep Learning

引用

1st international conference on Intelligent computing and Big data Analytics

作者： AbduRahim Barna, Rahul Kulkarni, Yashas Mydeen, Jamal Pulgam, Namita D. DY Patil Deemed Be Univ Ramrao Adik Inst Technol Dept Comp Engn Navi Mumbai India

ISBN: (纸本)9783031746819;9783031746826

In an era marked by the rapid proliferation and alteration of manipulated multimedia content, deepfake technology poses a significant threat to the authenticity of digital media. This media, which includes images and videos, necessitates rigorous evaluation for deepfake detection. To address this challenge, we propose a system called VeriFace, designed to accurately differentiate between authentic and fake digital content. VeriFace leverages advanced technologies such as face recognition, face masking, and eye blinking detection to provide a robust and scalable solution for real-time deepfake identification. The system is engineered with a user-friendly interface, ensuring accessibility and ease of use. Through its innovative approach, VeriFace aims to enhance the security and authenticity of digital content, offering a crucial tool for safeguarding information in both digital and real-world contexts.

关键词： Multimedia Deepfake Face Masking Eye Blinking Convolutional Neural Network Accuracy

来源：评论

学校读者我要写书评

暂无评论

Amend to Alignment: Decoupled Prompt Tuning for Mitigating Spurious Correlation in Vision-Language Models 41

Amend to Alignment: Decoupled Prompt Tuning for Mitigating S...

引用

41st international conference on machine learning, ICML 2024

作者： Zhang, Jie Ma, Xiaosong Guo, Song Li, Peng Xu, Wenchao Tang, Xueyang Hong, Zicong Department of Computing The Hong Kong Polytechnic University Hong Kong Department of Computer Science and Engineering The Hong Kong University of Science and Technology Hong Kong School of Cyber Science and Engineering Xi'an Jiaotong University Shaanxi China

Fine-tuning the learnable prompt for a pre-trained vision-language model (VLM), such as CLIP, has demonstrated exceptional efficiency in adapting to a broad range of downstream *** prompt tuning methods for VLMs do not distinguish spurious features introduced by biased training data from invariant features, and employ a uniform alignment process when adapting to unseen target *** can impair the cross-modal feature alignment when the test data significantly deviate from the distribution of the training data, resulting in a poor out-of-distribution (OOD) generalization *** this paper, we reveal that the prompt tuning failure in such OOD scenarios can be attribute to the undesired alignment between the textual and the spurious *** a solution, we propose CoOPood, a fine-grained prompt tuning method that can discern the causal features and deliberately align the text modality with the invariant ***, we design two independent contrastive phases using two lightweight projection layers during the alignment, each with different objectives: 1) pulling the text embedding closer to the invariant image embedding and 2) pushing the text embedding away from the spurious image *** have illustrated that CoOPood can serve as a general framework for VLMs and can be seamlessly integrated with existing prompt tuning *** experiments on various OOD datasets demonstrate the performance superiority over state-of-the-art methods. Copyright 2024 by the author(s)

关键词： Visual languages

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 << < 491 492 493 494 495 496 497 498 499 500 > >>

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：