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

575 篇 会议
15 册 图书
4 篇 期刊文献

馆藏范围

594 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

480 篇 工学
- 454 篇 计算机科学与技术...
- 127 篇 软件工程
- 85 篇 电气工程
- 45 篇 信息与通信工程
- 30 篇 控制科学与工程
- 15 篇 机械工程
- 15 篇 生物工程
- 10 篇 光学工程
- 7 篇 化学工程与技术
- 7 篇 航空宇航科学与技...
- 7 篇 生物医学工程（可授...
- 6 篇 建筑学
- 5 篇 土木工程
- 5 篇 安全科学与工程
- 4 篇 材料科学与工程（可...
153 篇 理学
- 113 篇 物理学
- 23 篇 数学
- 15 篇 生物学
- 11 篇 统计学（可授理学、...
- 5 篇 系统科学
- 4 篇 化学
40 篇 管理学
- 21 篇 管理科学与工程(可...
- 20 篇 图书情报与档案管...
- 6 篇 工商管理
15 篇 医学
- 9 篇 基础医学(可授医学...
- 9 篇 临床医学
14 篇 教育学
- 14 篇 教育学
13 篇 法学
- 7 篇 社会学
- 6 篇 法学
4 篇 经济学
- 4 篇 应用经济学
4 篇 农学
4 篇 军事学

主题

145 篇 machine learning
105 篇 artificial intel...
65 篇 deep learning
28 篇 adversarial mach...
24 篇 reinforcement le...
16 篇 object detection
15 篇 learning systems
14 篇 neural networks
14 篇 federated learni...
13 篇 computer vision
12 篇 cybersecurity
12 篇 security
12 篇 accuracy
12 篇 machine learning...
12 篇 training
11 篇 computer applica...
10 篇 multi-domain ope...
9 篇 ai
9 篇 explainable arti...
9 篇 unsupervised lea...

机构

6 篇 marwadi universi...
6 篇 national univers...
5 篇 us army engineer...
5 篇 us army res lab ...
4 篇 college of compu...
4 篇 siemens corporat...
4 篇 univ maryland co...
4 篇 carnegie mellon ...
4 篇 foundation for r...
3 篇 mcmaster univ 12...
3 篇 korea univ dept ...
3 篇 us mil acad army...
3 篇 ibm tj watson re...
2 篇 johns hopkins un...
2 篇 penn state univ ...
2 篇 the high school ...
2 篇 howard univ wash...
2 篇 wroclaw universi...
2 篇 univ southern ca...
2 篇 shanghai ai lab ...

作者

8 篇 rawat danda b.
6 篇 divyakant meva
6 篇 kalpesh popat
6 篇 sunil bajeja
6 篇 pankaj mudholkar
6 篇 sridaran rajagop...
4 篇 surucu onur
4 篇 giuliano alessan...
4 篇 alsadi naseem
4 篇 helmut degen
4 篇 hilal waleed
4 篇 stavroula ntoa
4 篇 raglin adrienne
4 篇 asher derrik e.
4 篇 yawney john
4 篇 markowitz jared
3 篇 bastian nathanie...
3 篇 schutte klamer
3 篇 kwak heon-gyu
3 篇 price stanton r.

语言

563 篇 英文
31 篇 其他
2 篇 中文

检索条件"任意字段=Conference on Artificial Intelligence and Machine Learning in Defense Applications IV"

共 594 条记录，以下是191-200 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：

Deducing 15cm detail from 30cm satellite images with Deep learning 4

Deducing 15cm detail from 30cm satellite images with Deep Le...

引用

conference on artificial intelligence and machine learning for Multi-Domain Operations applications iv

作者： Zhang, Bingcai Luu, Yen Vaidyanathan, Kalyan Paderes, Fidel Settergren, Reuben de Venecia, Kurt BAE Syst 10920 Technol Dr San Diego CA 92127 USA

ISBN: (纸本)9781510651036;9781510651029

We explore the application of single image super-resolution technique to satellite image and its effect on object detection performance. This technique uses a deep convolutional neural network to learn transformations between different zoom levels of image pyramids, also referred to as Resolution Set (Rset). The network can learn the transformations from the 2:1 RSet at a Ground Sample Distance (GSD) of 60cm to the full resolution image at a GSD of 30cm by minimizing the differences between ground-truth full resolution and the derived 2x zoom. After training, the learned transformation is applied to the 1:1 full resolution image transforming the pixels to 2x resolution. The learned transformations has intelligence built in and can infer higher resolution images. We find super-resolution images significantly improve object detection accuracy, improve manual feature extraction accuracy, and also benefit imagery analysis workflows and derived products which use satellite images.

关键词： super-resolution satellite image convolutional neural network image transformation object detection

来源：评论

学校读者我要写书评

暂无评论

Enhancing Inference Time and Memory Utilization for machine learning in Resource Constrained Internet of Battlefield Things 4

Enhancing Inference Time and Memory Utilization for Machine ...

引用

conference on artificial intelligence and machine learning for Multi-Domain Operations applications iv

作者： Ghimire, Bimal Rawat, Danda B. Howard Univ DoD Ctr Excellence AI ML Dept Elect Engn & Comp Sci Washington DC 20059 USA

ISBN: (纸本)9781510651036;9781510651029

Convolution Neural Network (CNN) models have demonstrated remarkable success for many applications including computer vision in recent years. However, the increase in accuracy achieved so far is at the cost of memory and complex computations. The CNN models are growing deeper and wider making it difficult to be fit in a single device that has limited resources. Moreover, the inference time of such models is large enough to be unfit to apply for the real-time mission critical applications like the Internet of Battlefield Things (IoBT). In IoBT, where the unmanned aerial vehicles (UAVs) flying in the battlefield zone and capturing images, require accurate learning and immediate inference. It becomes problematic if the learning model does not fit in a single UAV when it has limited resources. Considering the aforementioned issues, in this paper, we study a formal approach to improve the inference time and memory utilization in resource constrained IoBT. We consider that multiple UAVs are involved in the inference process in which we apply spatially parallel convolution and pooling operations for all the convolution layers and pooling layers as well as model parallelism for fully connected (FC) layers. Finally, we present the numerical results for varying number of participating UAVs, input data/image sizes, and communication speeds.

关键词： Convolution Neural Network Computer Vision Internet of Battlefield Unmanned Aerial Vehicles Spatially Parallel Convolution Model Parallelism

来源：评论

学校读者我要写书评

暂无评论

Block-term Dropout For Robust Adversarial defense 34

Block-term Dropout For Robust Adversarial Defense

引用

34th IEEE International conference on Tools with artificial intelligence (ICTAI)

作者： Ouyang, Chen Yang, Wei Hu, Rong Univ Sci & Technol China Sch Comp Sci & Technol Hefei Peoples R China

ISBN: (纸本)9798350397444

Deep neural networks (DNNs) have lately shown tremendous performance in various applications. However, along-side their superiority in these tasks, recent studies have demonstrated that DNNs are easily fooled by adversarial attacks. To guard against adversarial examples, we provide a new solution to hardening DNNs through Block-term Dropout (BT-Dropout), an adversarial defense technique that leverages a latent high-order factorization of the network. Specifically, we impose low-rank block-term tensor structure on the weights of fully-connected layer to obtain compact networks, and then apply BT-Dropout in the latent subspace without pruning the weights directly. Meanwhile, for activation tensor fed into fully-connected layer, Tucker Dropout which can be viewed as a special case of BT-Dropout is introduced to preserve multilinear structure of activations. Furthermore, we show that BT-Dropout implicitly regularizes the tensor decomposition. Comprehensive experiments have demonstrated the effectiveness of our proposed method to improve the adversarial robustness for the models on standard image classification benchmarks.

关键词： Adversarial machine learning Tensor Decomposition Dropout Deep Neural Networks

来源：评论

学校读者我要写书评

暂无评论

Defending Against Poisoning Attacks in Federated Prototype learning on Non-IID Data 18th

Defending Against Poisoning Attacks in Federated Prototype ...

引用

18th International conference on Wireless artificial Intelligent Computing Systems and applications, WASA 2024

作者： Zhang, Jie Zhang, Hongliang Wang, Guijuan Dong, Anming Jinan250353 China Shandong Provincial Key Laboratory of Industrial Network and Information System Security Shandong Fundamental Research Center for Computer Science Jinan250353 China

ISBN: (纸本)9783031714665

Federated learning (FL) is an emerging distributed machine learning paradigm that enables participants to cooperatively train learning tasks without revealing the raw data. However, the distributed nature of FL makes it susceptible to poisoning attacks, especially when the local data of participants are non-independent and identically diatributed (non-IID). Although several defense methods have been proposed to mitigate poisoning attacks, their effectiveness is limited by the specific assumptions about the data distribution. In this work, we propose a new defense strategy, FedAPA (Federated Prototype learning Against Poisoning Attacks). Specifically, we use abstract class prototypes to communicate between the clients and server, thus effectively alleviating the impact of non-IID data. Moreover, we propose a new abnormal client detection method that aims to mitigate the impact of malicious clients while distinguishing between malicious and benign clients, thereby effectively defending against poisoning attacks. Extensive experiments on different datasets show that FedAPA can effectively resist the poisoning attacks under various data distributions. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

A Federated Parameter Aggregation Method for Node Classification Tasks with Different Graph Network Structures 2nd

A Federated Parameter Aggregation Method for Node Classific...

引用

2nd International conference on Data Security and Privacy Protection, DSPP 2024

作者： Song, Hao Yao, Jiacheng Chen, Xuanze Xu, Shaocong Jin, Shibo Zhou, Jiajun Fu, Chenbo Xuan, Qi Yu, Shanqing Institute of Cyberspace Security Zhejiang University of Technology Hangzhou310023 China Binjiang Cyberspace Security Institute of ZJUT Hangzhou310056 China College of Information Engineering Zhejiang University of Technology Hangzhou310023 China

ISBN: (纸本)9789819785391

Over the past few years, federated learning has become widely used in various classical machine learning fields because of its collaborative ability to train data from multiple sources without compromising privacy. However, in the area of graph neural networks, the nodes and network structures of graphs held by clients are different in many practical applications, and the aggregation method that directly shares model gradients cannot be directly applied to this scenario. Therefore, this work proposes a federated aggregation method FLGNN applied to various graph federation scenarios and investigates the aggregation effect of parameter sharing at each layer of the graph neural network model. The effectiveness of the federated aggregation method FLGNN is verified by experiments on real datasets. Additionally, for the privacy security of FLGNN, this paper designs membership inference attack experiments and differential privacy defense experiments. The results show that FLGNN performs good robustness, and the success rate of privacy theft is further reduced by adding differential privacy defense methods. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Secondary Students' Emerging Conceptions of AI: Understanding AI applications, Models, Engines and Implications 24

Secondary Students' Emerging Conceptions of AI: Understandin...

引用

6th conference on United Kingdom and Ireland Computing Education Research

作者： Whyte, Robert Kirby, Diana Sentance, Sue Univ Cambridge Cambridge England

ISBN: (纸本)9798400711770

Despite the growing use of AI technologies, there is limited empirical evidence of secondary students' understanding of AI. Gaining insights into ways in which this might be influenced by their everyday engagement with these tools can support the teaching of AI, including the design of educational resources. The SEAME framework proposes four dimensions for thinking about AI: (i) the social and ethical (SE) implications of AI;(ii) its use in applications (A);(iii) how models (M) are trained;and (iv) the underlying engines (E) used. In this paper, we investigate students' emerging conceptions of AI and how these can be understood through classification using an appropriate framework. Drawing on data from 474 secondary and college students (11-18), we found that students mostly focused on the applications of AI, followed by its social and ethical implications. We found evidence that students held some primarily accurate conceptions of how AI works, such as how models are trained using large datasets and how human behaviours are simulated (e.g. talking, problem-solving). Specific conceptions relating to generative AI tools were also observed, including the ability to generate content and response to prompts. Conversely, we found students held naive conceptions relating to AI systems having agency or emotions. We argue that efforts to promote accurate conceptions of AI should build on students' conceptions and take into account how language and representation are used (e.g. anthropomorphism). The results of this study have implications for educators, resource developers and researchers.

关键词： computing education K-12 education artificial intelligence machine learning conceptions

来源：评论

学校读者我要写书评

暂无评论

Wordle-Based Text Analysis: applications of Intelligent Algorithms Such as XGBoost and Char-CNN 3rd

Wordle-Based Text Analysis: Applications of Intelligent Algo...

引用

3rd International conference on machine learning, Cloud Computing and Intelligent Mining, MLCCIM 2024

作者： Ye, Huicheng Su, Qifeng Chen, Shuang Xu, Dan Yu, Hongqi National University of Defense Technology Hunan Changsha410000 China School of Systems Science and Engineering Sun Yat-sen University Guangzhou510006 China

ISBN: (纸本)9789819624676

In this paper, the application of intelligent algorithms such as XGBoost and Char-CNN is verified by establishing the analysis model of the game Wordle. The three models established can perform the tasks of fitting and predicting the number of players, analyzing the number of attempts, and classifying the difficulty of the game. In the first model, XGBoost optimizes the classical SIR Model to improve output accuracy and analysis efficiency. In the second model, this paper designs an innovative two-dimensional code word coding method concerning machine language, which effectively reduces the redundancy of Char-CNN. In the third model, the classical Bayesian algorithm is used to classify the difficulty of a large word set, proving its usability. In addition to helping Wordle editors identify more appealing puzzles, the process of analyzing text with intelligent algorithms is also valuable for scholars who study the text features. In the era of big data, the model-building method in this paper may also give some enlightenment to the field of artificial intelligence. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Char-CNN Wordle XGBoost

来源：评论

学校读者我要写书评

暂无评论

Towards artificial intelligence Empowered Security and Privacy Issues in 6G Communications

Towards Artificial Intelligence Empowered Security and Priva...

引用

2022 International conference on Sustainable Computing and Data Communication Systems, ICSCDS 2022

作者： Bandi, Ajay Yalamarthi, Sowmya School of Computer Science and Information Systems Northwest Missouri State University Maryville United States

ISBN: (纸本)9781665478847

applications of wireless networks beyond 5G are vulnerable to various security and privacy concerns. This research aims to identify the security and privacy flaws beyond 5G network applications and their defense mechanisms. This research study has reviewed 44 research articles and presented the taxonomy of several concerns in security and privacy of artificial intelligence and machine learning empowered 6G applications. Furthermore, this study has identified the defense mechanism technologies for such issues. A few defense mechanism approaches are homographic encryption, physical layer authentication, biometric authentication, explainable, trustworthy, and superintelligent AI. The implications of this study are applicable to prevent attacks, eavesdropping, jamming, and other security and privacy concerns when developing wireless networks. © 2022 IEEE.

关键词： Wireless networks

来源：评论

学校读者我要写书评

暂无评论

Survey on artificial intelligence and Its applications 30th

Survey on Artificial Intelligence and Its Applications

引用

Advances in Systems Engineering - Proceedings of the 30th International conference on Systems Engineering, ICSEng 2023

作者： Goel, Sarthak Guha, Anamika Uma, Kuppusamy Shanmugam, Thirumurugan School of Information Technology & Engineering Vellore Institute Technology Vellore632014 India Department of Information Technology University of Applied Sciences Suhar Sohar Oman

ISBN: (纸本)9783031405785

Be self-driving cars, multi-media recognition and processing, or content recommendation, artificial intelligence has wide range of applications in mitigating climate change, food shortage, defense equipment designing, cybersecurity, creating advance simulations supply chain, pollution, space exploration, genetic engineering, geology, and much more. Often termed as a bane, ethical AI is certainly impacting the world positively because of its uncontrolled potential. The ability to learn from past experiences enables machines and devices to perform tasks in complex, real-world situations, without constant user guidance. In this research, we propose a discussion on recent trends and advancements made in the field of Education. We deliver case studies of how artificial intelligence and its sub-domains like Natural Language Processing, Computer Vision, and machine learning, along with sister technology like Augmented Reality and Virtual Reality, Cloud Computing, Brain Computing, IoT data processing, and Data Science are working hand in hand to provide significant up-gradation to the whole education sector. The research talks about artificial intelligence’s impact on the teaching and administrative aspects of the education sector. We dive deep into the benefits generated on student’s and faculty’s sides while touching upon how artificial intelligence has speeded up the administrative processes such as student selection, reports generation, resource planning, curriculum generation and analysis, and much more. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Big data

来源：评论

学校读者我要写书评

暂无评论

Neural Network Training Loss Optimization Utilizing the Sliding Innovation Filter 4

Neural Network Training Loss Optimization Utilizing the Slid...

引用

conference on artificial intelligence and machine learning for Multi-Domain Operations applications iv

作者： Alsadi, Naseem Hilal, Waleed Surucu, Onur Giuliano, Alessandro Gadsden, Stephen A. Yawney, John AlShabi, Mohammad McMaster Univ 1280 Main St W Hamilton ON L8S 4L8 Canada Adastra Corp 8500 Leslie St 600 Thornhill ON L3T 7M8 Canada Univ Sharjah Sharjah U Arab Emirates

ISBN: (纸本)9781510651036;9781510651029

artificial feedforward neural networks (ANN) have been traditionally trained by backpropagation algorithms involving gradient descent algorithms. This is in order to optimize the network's weights and parameters in the training phase to minimize the out of sample error in the output during testing. However, gradient descent (GD) has been proven to be slow and computationally inefficient in comparison with studies implementing the extended Kalman filter (EKF) and unscented Kalman filter (UKF) as optimizers in ANNs. In this paper, a new method of training ANNs is proposed utilizing the sliding innovation filter (SIF). The SIF by Gadsden et al. has demonstrated to be a more robust predictor-corrector than the Kalman filters, especially in ill-conditioned situations or the presence of modelling uncertainties. In this paper, we propose implementing the SIF as an optimizer for training ANNs. The ANN proposed is trained with the SIF to predict the MackeyGlass Chaotic series, and results demonstrate that the proposed method results in improved computation time compared to current estimation strategies for training ANNs while achieving results comparable to a UKF-trained neural network.

关键词： Optimization Estimation Theory Kalman Filter Gradient Descent Smooth Variable Structure Filter Sliding Modes State-Space Methods Deep learning machine learning Neural Networks

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60页 << < 16 17 18 19 20 21 22 23 24 25 > >>

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：