检索结果-内蒙古大学图书馆

IoT Based Control Networks and Intelligent Systems (ICICNIS), International Conference on

作者： Venkateshwari Subburaj Pandiselvam Pandiyarajan Department of Computer Application School of Computing Kalasalingam Academy of Research and Education Krishnankoil TamilNadu India Department of Computer Science and Engineering School of Computing Kalasalingam Academy of Research and Education Krishnankoil TamilNadu India

ISBN: (数字)9798331518097

ISBN: (纸本)9798331518103

Fetal brain anomaly prediction is important for fetal medicine, as well as for prenatal health care. Fetal anomalies are classified into two types, anomalies in the fetus' body parts including heart, lung, and kidney anomalies in facial features. Thus, to predict the anomalies in the fetal brain, various deep-learning models are used in this paper. The LeNet, AlexNet, ResNet, VGGNet, GoogLeNet and ZFNet were used. From the results obtained, the ResNet method achieved 99% accuracy with the given 3D ultrasound images of the fetal brain compared with other models and benchmark algorithms like SVM, Logistic regression, and KNN. The results of this paper are useful to identify fetal anomalies. Hence, this approach will be useful for improving outcomes for individuals and society as much as possible.

关键词： Deep learning Support vector machines Solid modeling Pediatrics Three-dimensional displays Accuracy Ultrasonic imaging Predictive models Brain modeling Prediction algorithms

来源：评论

学校读者我要写书评

暂无评论

A Reinforcement Learning-Based Image Mapping-Topology Optimization for Antenna Design

引用

IEEE Antennas and Wireless Propagation Letters 2025年

作者： Dou, Jiangling Gong, Hao Chen, Haokang Chen, Yujie Shen, Tao Song, Jian Kunming University of Science and Technology Faculty of Information Engineering and Automation Yunnan Key Laboratory of Computer Technologies Application Kunming650500 China Kunming University of Science and Technology Faculty of Information Engineering and Automation Kunming Yunnan650500 China College of Mechanical and Electrical Engineering Yunnan Electromechanical Vocational and Technical College China

A Reinforcement Learning (RL)-based Image Mapping Topology (IMT) optimization approach is proposed. To enable RL to optimize antenna topology and increase the efficiency of the optimization process. Two crucial components, Stratified Sampling Balance (SSB) and IMT are introduced. SSB is utilized to balance different types of data samples. IMT is utilized to facilitate the interaction between RL and simulation software, it also pretrains the actor network within the RL algorithm. This methodology allows the RL agent to acquire knowledge from the interaction of the antenna topology with the environment. Subsequently, the agent outputs antenna topology with the maximum reward signal. Compared with other machine learning-based optimization methods, the proposed method significantly alleviates the issue of learning inefficiency and reduces the exploration time for feasible solutions. To validate the effectiveness of the optimization method, a microstrip antenna is optimized. The measurement results show that the impedance bandwidth is doubled while maintaining the realized gain with fewer simulation samples and less optimization time. © 2025 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Timeliness-Selective Incentive Federated Crowdsourcing

Timeliness-Selective Incentive Federated Crowdsourcing

引用

IEEE International Conference on Web Services (ICWS)

作者： Xiaoqian Jiang Haiyang Diao Cangqi Zhou Jing Zhang School of Cyber Science and Engineering Southeast University China Siemens (China) Co. Ltd. School of Computer Science and Engineering Nanjing University of Science and Technology China Engineering Research Center of Blockchain Application Supervision and Management (Southeast University) Ministry of Education China

ISBN: (数字)9798350368550

ISBN: (纸本)9798350368567

Crowdsourcing offers an efficient means of gathering labeled data and training machine learning models. However, the exposure of crowd workers to potential privacy breaches during data collection diminishes their willingness to participate. This paper proposes TsIFedCrowd, a novel incentive mechanism within the federated learning paradigm, aimed at mitigating privacy risks for crowd workers while achieving high-quality crowdsourced data and learning models at minimal expense. TsIFedCrowd enables clients to retain privacy-sensitive data locally, requiring only the upload of trained models to the server, which aggregates them into a global model. By modeling the federated crowdsourcing process as a two-stage Stackelberg game, TsIFedCrowd incentivizes workers to complete tasks with heightened quality and efficiency. It also provides a timeliness selection mechanism, which makes TsIFedCrowd suitable for both real-time crowdsourcing, where the generation time of task results (data) needs to meet the established requirements of the application, and non-real-time crowdsourcing. TsIFedCrowd mandates clients to complete federated crowdsourcing tasks while maximizing utility for both clients and the server, achieved through Nash equilibrium resolution. TsIFedCrowd is also extended to the multiple heterogeneous federated crowdsourcing scenario. Extensive experiments demonstrate TsIFedCrowd’s effectiveness in ensuring incentive fairness and universality, accelerating convergence speed, and budget savings. Real-world crowdsourcing tasks further validate its efficacy.

关键词： Crowdsourcing Training Web services Scalability Games Nash equilibrium Privacy breach Data models Real-time systems Servers

来源：评论

学校读者我要写书评

暂无评论

Design of data processing scheme of 5G terminal protocol conformance test system

Design of data processing scheme of 5G terminal protocol con...

引用

Advances in Electrical, Computing, Communication and Sustainable Technologies (ICAECT), International Conference on

作者： Cao Jingyao Amit Yadav Asif Khan Abhishek Pratap Sah School of Computer and Software Chengdu Neusoft University Chengdu China College of Engineering IT and Environment Charles Darwin University NT Australia Department of Computer Application Integral University Lucknow India School of Management Science and Engineering University of Electronic Science and Technology of China Chengdu China

Considering the fact that the test case runs abnormally and the test case execution fails during the development and debugging of the 5G terminal protocol conformance test system. Here useful approach is implemented by embedding data copy and storage module in the system, which completes the encapsulation, copy and storage of massive data generated during the operation of the system, so that equipment developers can correctly locate the errors in the system according to the data, and repair the system loopholes to ensure the stability and accuracy of the system.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Music Generation using Time Distributed Dense Stateful Char-RNNs 7

Music Generation using Time Distributed Dense Stateful Char-...

引用

7th IEEE International conference for Convergence in Technology, I2CT 2022

作者： Banerjee, Shobhan Rath, Manas Swain, Tanmaya Samant, Tapaswini Birla Institute of Technology and Science - Pilani Rajasthan India Kalinga Institute of Industrial Technology School of Computer Application Bhubaneswar India Kalinga Institute of Industrial Technology School of Electronics Engineering Bhubaneswar India

ISBN: (纸本)9781665421683

Sequence generation is one of the state-of-the-art topics in recent days, where given a sequence of inputs, the aim is to generate a similar sequence of outputs in a given context. The applications range from sentence autocompletion in mail bodies to text suggestions for automatic replies, etc. A similar idea can therefore be utilized to generate a sequence of musical notes using some LSTM/GRU based architecture, where we train our model based on given sequences of musical notes. If we look at music of a certain genre, it's simply a time series data where the data is in frequency domain. Fortunately for musicians these frequencies have a notation, which can serve as texts to train any given time series model. Hence, in this paper we propose a Char-RNN based model which can understand the patterns in each composition or a raga and generate new piece of music based on that. The model must not simply copy paste the sequence or generate any random note at a given instant of time but be capable enough to grasp the patterns in which the given piece of music is based upon and create a similar, new piece of music out of that. © 2022 IEEE.

关键词： Music

来源：评论

学校读者我要写书评

暂无评论

Cascade Temporal Convolutional Network for Multitask Learning

Cascade Temporal Convolutional Network for Multitask Learnin...

引用

International Conference on Artificial Intelligence and Big Data (ICAIBD)

作者： Huiyun Zhang Heming Huang Henry Han Department of Computer Science School of Engineering and Computer Science Baylor University USA School of Computer Science Qinghai Normal University Xining China School of Computer Science Qinghai Normal University The State Key Laboratory of Tibetan Intelligent Information Processing and Application Xining China

Multi-task learning (MTL) is a powerful technique in machine learning that enables the simultaneous training of multiple tasks. In this paper, we propose a novel MTL approach for the recognition of emotions, speakers, and gender in speech signals. Our approach presents a novel deep learning model called the Cascade Temporal Convolutional Network (CTCN to handle MTL efficiently. CTCN is designed to effectively improve the parallel processing ability of the model in multitasking learning. It provides an effective solution for improving the performance of deep learning in multitasking scenarios. CTCN is particularly adept at processing sequences of data with varying lengths and extracting features that are relevant to multiple tasks. Our experimental results demonstrate that the proposed CTCN approach for MTL achieves significantly superior performance on all tasks on benchmark datasets compared to other deep learning models. This study will inspire more future work in the MTL field.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient Human Detection in Real- Time Using Vision Transformers: Balancing Performance and Latency for Security and Autonomous applications

Efficient Human Detection in Real- Time Using Vision Transfo...

引用

Next Generation Communication & Information Processing (INCIP), International Conference on

作者： Jeremiah Ocheni Immanuel Johnraja Jebadurai Getzi Jeba Leelipushpam Paulraj Difference Finbarrs Dept. of Computer Science and Engineering Karunya Institute of Technology and Sciences Coimbatore Tamil Nadu India Dept. of Business Admin and Computer Application Rathinam College of Arts and Sciences Coimbatore Tamil Nadu India

ISBN: (数字)9798331528140

ISBN: (纸本)9798331528157

With applications in security and surveillance, au-tonomous driving, human-computer interaction, the application of Vision Transformers (ViTs) is a major breakthrough in deep learning techniques, outperforming more conventional methods such as Convolutional Neural Networks (CNNs). Human detection is an essential component of computer vision. Using a custom dataset, this paper analyses the use of Vision Transformers (ViT) for human detection. We introduce a thorough approach to training and evaluating models that makes use of PyTorch. To strike a balance between the computational efficiency and performance with balancing accuracy, and the latency as part of the study to evaluate the effectiveness of the approach in real-time applications. The findings show that Vision transformer model with a cross-entropy loss function and the Adam optimizer can attain competitive performance in terms of accuracy 84% and minimal latency of 0.11 secs, and Equal Error Rate of 0.1845 indicating that it has potential uses in human detection and can be use in the real time applications.

关键词： Deep learning computer vision Visualization Accuracy Computational modeling Transformers Real-time systems Computational efficiency Security Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A Comparative Analysis of Cloud based Services Platform 6

A Comparative Analysis of Cloud based Services Platform

引用

6th Smart Cities Symposium, SCS 2022

作者： Bajaj, Karan Sharma, Bhisham Singh, Raman Kumar, Mukesh Chowdhury, Subrata Department of Computer Science & Engineering Chitkara University School of Engineering and Technology Chitkara University Himachal Pradesh India School of Computing Engineering and Physical Sciences University of the West of Scotland Lanarkshire United Kingdom School of Computer Application Lovely Professional University Punjab Phagwara India MCA Department Sri Venkateswara College of Engineering and Technology India

ISBN: (纸本)9781839538544

Data processing at high speeds and storage solutions on a massive scale are two things that are becoming increasingly important as a result of the proliferation of new start-up companies and the vast volume of data that users consume daily. The cloud platforms offer a solution to these issues in the shape of virtualization, which entails the creation of several distinct virtual machines within a single actual system. Because of this, the efficiency of the processor increases, and it now has far less downtime. The four major cloud platforms are discussed, and each has its own set of benefits, which helps make them successful in their unique conducts. Although Amazon EC2 is one of the most used and earliest that offers assistance for the utmost numeral of already configured system software termed as operating systems, which falls short in terms of both its scope and the support that it makes available. In a similar vein, the Google Cloud Platform can support the greatest number of databases and has an excellent repository of in-built libraries. The Microsoft Azure has the greatest number of users out of four, however, it is badly lacking in the database supports it offers. IBM Watson offers user friendly services but has scalability issues. As a result, a conclusion has been reached, that the selecting a particular cloud platform, there is extreme dependence on the needs of the users. These requirements may vary from client to client which are the end-users of the applications. Before deciding on a particular platform, it is best to conduct a thorough analysis of the needs. © 2022 IET Conference Proceedings. All rights reserved.

关键词： Digital storage

来源：评论

学校读者我要写书评

暂无评论

The Open Source Threat Intelligence Relational Dataset and Its Optimal Implementation

The Open Source Threat Intelligence Relational Dataset and I...

引用

Innovation in Technology (APCIT), Asia Pacific Conference on

作者： Yaru Yang Junyu Li Ronghua Zhang Abhishek Pratap Sah Amit Yadav Asif Khan Department of Computer and Software Chengdu Neusoft University Chengdu China School of Management Science and Engineering University of Electronic Science and Technology of China Chengdu China College of Engineering IT and Environment Charles Darwin University NT Australia Department of Computer Application Integral University Lucknow India

ISBN: (数字)9798350361537

ISBN: (纸本)9798350361544

Threat intelligence provides a platform for cybersecurity engineers for attack traceability, which provides substantial knowledge database logs to defend against future security threats. Threat intelligence relationship extraction based on deep learning solves the challenge of threat knowledge construction to a certain extent but still faces problems such as lack of open-source datasets and the inability of the model to accurately correlate threat entities with potential relationships. Therefore, for cybersecurity research work, this paper designs a threat ontology, constructs the threat relationship dataset TreatRE by remote supervision, and opens this dataset. The dataset contains 12000 utterances and 12 threat relations from 500 CTIs, and it performs well in multiple relation models trained on deep learning methods. Meanwhile, we propose a multisensory attention-based threat intelligence relationship extraction method MAtt, which combines location perception, self-attention perception, and neuronal memory perception to further improve the threat relationship extraction effect. Experimental results show that the trained model based on TreatRE can more accurately extract the knowledge objects and their relationships described in threat intelligence. An accuracy score of 95.4% can be obtained using the MAtt method, which is 3.48% more than the best baseline compared with the same type of relationship extraction model.

关键词： Deep learning Knowledge engineering Technological innovation Databases Buildings Asia Ontologies

来源：评论

学校读者我要写书评

暂无评论

Stream Attention Based U-Net for L3DAS23 Challenge 48

Stream Attention Based U-Net for L3DAS23 Challenge

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Wang, Honglong Fu, Yanjie Li, Junjie Ge, Meng Wang, Longbiao Qian, Xinyuan Tianjin University Tianjin Key Laboratory of Cognitive Computing and Application College of Intelligence and Computing Tianjin China The Chinese University of Hong Kong Shenzhen China Co. Ltd Tianjin China University of Science and Technology Beijing School of Computer and Communication Engineering Beijing China

ISBN: (纸本)9781728163277

Machine learning applications of 3D audio are gaining increasing interest in recent years. In this paper, we propose a stream attention based U-Net to remove background noise and reverberation based on ICASSP Signal Processing Grand Challenge 2023: L3DAS23 Challenge1 Audio-only track task1 3D Speech Enhancement. Results show that proposed method achieves superior performance than the official baseline model. © 2023 IEEE.

关键词： 3D Speech enhancement L3DAS23 Challenge

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：