检索结果-内蒙古大学图书馆

SSRN 2024年

作者： Zhang, Jianhua Gao, Yi Zhou, Mian Liu, Ruyu Cheng, Xu Nikolić, Saša V. Chen, Shengyong School of Computer Science and Engineering Tianjin University of Technology Tianjin300384 China School of Information Science and Technology Hangzhou Normal University Hangzhou311121 China Department of Electronics Faculty of Electronic Engineering University of Nis Nis Serbia

Recent advancement of Knowledge distillation (KD) is to extract and transfer middle layer knowledge of teacher models to student models, which is better than original KDs which only transfer the last layer of knowledge. However, the middle layer knowledge commonly appears as a complex tensor, which is more difficult to transfer than the one-dimensional knowledge in the last layer. Moreover, when there are large differences between the teachers and students in terms of model parameter capabilities and model structures, the middle layers of teacher and student models will not be aligned, which will further increase the learning difficulty for student models. To solve these problems, we propose a novel knowledge extraction module to transform the complex tensor-based knowledge in middle layers to one-dimensional knowledge based on Singular Value Decomposition. Thus, the knowledge at the middle layers of teacher models can be effectively extracted and simplified, and extremely facilitate the learning of student models, even though the structure and parameter capacity between the teacher network and the student network are very different. To help the students learn the knowledge from the middle layers of teachers as accurately as possible, we also propose a novel loss function that can constrain values of the one-dimension knowledge learned by the student model to be as close as possible to that extracted from the teacher model, thus improving the learning efficiency of the student model. We have conducted extensive experiments on three major datasets (CIFAR-10, CIFAR-100, and ImageNet), and the results of our method demonstrate superior performance by comparing with the state-of-the-art methods. © 2024, The Authors. All rights reserved.

关键词： Singular value decomposition

来源：评论

学校读者我要写书评

暂无评论

Exploring the Boundaries of Real- Time Data Analysis with Machine Learning

Exploring the Boundaries of Real- Time Data Analysis with Ma...

引用

Communication & Computing (WCONF), World Conference on

作者： Bhavesh Agnihotri Balakumar P Nagaraju Bogiri B Reddy Asha. KS Dhvan Yadav Department of Computer Science & Engineering Vivekananda Global University Jaipur India Department of Information Technology Prince Shri Venkateshwara Padmavathy Engineering College Chennai-127 Department of Computer Engineering Vishwakarma Institute of Information Technology Pune INDIA Centre of Interdisciplinary Research in Business and Technology Chitkara University Institute of Engineering and technology Chitkara University Punjab Department of Electronics and Communication Engineering Faculty of Engineering and Technology Bangalore Karnataka India Maharishi School of Engineering and Technology Maharishi University of Information Technology Uttar Pradesh India

ISBN: (数字)9798350395327

ISBN: (纸本)9798350395334

Gadget studying (ML) is becoming more and more famous for real-time facts analysis responsibilities; however, present ML models are often no longer suitable for this form of software due to constraints on computational assets, facts distribution, or modeling complexity. In this paper, we investigate the limits of actual-time statistics evaluation for ML fashions to permit us to extract insights from massive datasets with constrained sources., we focus on key challenges: first, designing ML algorithms that require minimum pre-processing and make system information speedy; and second, locating powerful ways of calculating the accuracy of predictions made in real time. To address those troubles, we increase a framework for real-time information evaluation primarily based on current advances in extreme gaining knowledge of machines (ELM), which is a computationally efficient ML set of rules. We then examine the performance of our framework on a ramification of datasets and display that our ELM-based framework is capable of presenting correct predictions with low computational complexity and limited facts pre-processing. Our outcomes display that it's far viable to gain short, correct real-time information evaluation with ML models, and propose that, in addition, studies into ELM and other green ML algorithms could allow us to push the limits of real-time records analysis even similarly.

关键词： Analytical models Privacy Accuracy Computational modeling Scalability Software algorithms Prediction algorithms Market research Real-time systems Software

来源：评论

学校读者我要写书评

暂无评论

Improving Coverage Efficiency with Autonomous and Adaptive Communications Systems 15

Improving Coverage Efficiency with Autonomous and Adaptive C...

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Giri, Rakesh Kumar Mahajan, Vibhor Narula, Shaweta Selvi, S. Dhabliya, Dharmesh Putra, Riyan Sisiawan Sunrise University Department of Computer Science & Engineering College Alwar301028 India Chitkara University Centre of Research Impact and Outcome Punjab Rajpura140417 India Nutan College of Engineering and Research Department of Electronics and Telecommunication Pune410507 India Faculty of Management Studies Cms Business School India Vishwakarma Institute of Technology Department of Information Technology Maharashtra Pune India College Universitas Nahdlatul Ulama Surabaya Department of Management Surabaya60222 Indonesia

ISBN: (纸本)9798350370249

This paper presents a method to improve coverage efficiency in Wi-Fi communication systems by leveraging self-sustaining and adaptive communication systems. The approach seeks to exploit machine getting-to-know and related optimization techniques to enable network nodes to autonomously adjust electricity stages, pick out adjacent nodes, regulate transmission stages, and other parameters to acquire progressed insurance and higher record rates. Self-reliance getting to know an algorithm primarily based on reinforcement gaining knowledge of is used to pattern the surroundings, optimizes variables, and permit gadget adaptation even as minimizing human intervention. The paper discusses the structure and an algorithm evolved and explores the blessings of self-sufficient and adaptive communication systems over conventional non-autonomous systems. © 2024 IEEE.

关键词： Adjacent Minimizing Primarily Stages Transmission

来源：评论

学校读者我要写书评

暂无评论

Deep Neural Networks for Halyomorpha Halys Detection

Deep Neural Networks for Halyomorpha Halys Detection

引用

Mediterranean Conference on Control and Automation (MED)

作者： Alexandru Dinca Nicoleta Angelescu Loretta Ichim Dan Popescu Faculty of Automatic Control and Computer Science University Politehnica of Bucharest Bucharest Romania Faculty of Electrical Engineering Electronics and Information Technology Valahia University of Targoviste Targoviste Romania

Pest detection and identification in a timely manner is a crucial step for precision agriculture. Halyomorpha Halys is a common pest whose negative effects are known in agricultural areas and on various crops. The present work implemented and studied four performant neural networks, VGG19_BN, EfficientNetB7, DenseNet161, and ResNet152 for the detection of these insects. Although the detection of these insects in the natural environment through automated means, excluding traps, is a challenge, the results obtained are promising.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Quasi FSS Reflector with Dualband Cross Dipole for 5G Base Station Antenna Application

A Quasi FSS Reflector with Dualband Cross Dipole for 5G Base...

引用

International Electrical engineering Congress (iEECON)

作者： Wisit Loedhammacakra Maleeya Tangjitjetsada Chawalit Rakluea Norakamon Wongsin Thinnawat Jangjing Nonchanutt Chudpooti Chatree Mahatthanajatuphat Tawan Thianthong Department of Electronics and Telecommunication Engineering Faculty of Engineering Rajamangala University of Technology Thanyaburi Pathum Thani Thailand Department of Electronics and Telecommunication Engineering Faculty of Engineering and Architecture Rajamangala University of Technology Suvarnabhumi Nonthaburi Thailand Department of Electrical and Computer Engineering Faculty of Engineering King Mongkut's University of Technology North Bangkok Bangkok Thailand Siam Airport Ground Services Co. Ltd. Bangkok Thailand

This article describes the application of a quasi-frequency selective surface (FSS) reflector to a cross-dipole antenna for a 5G base station communication system. The quasi-FSS is composed of a square patch resonator with a resonance frequency of 1800 MHz and a perfect electric conductor (PEC) that reflects electromagnetic waves of all frequencies. Moreover, it can support and reflect two linear polarizations (±45°) of a dual-band cross-dipole antenna with dual operating frequencies of 700 MHz and 1800 MHz. At 700 MHz and 1800 MHz, the impedance characteristic of the antenna with reflector is approximately 26.83% and 34.86 % for Sll and S22 ≤-10 dB (±45°). The measured peak gain of the antenna is around 5.74 dBi at 775MHz and 11.88 dBi at 1850MHz. With a quasi-FSS reflector, the proposed antenna can also create a directional radiation pattern and facilitate the utilization of a 5G mobile communication base station.

关键词： Base stations Dipole antennas 5G mobile communication Communication systems Resonant frequency Dual band Mobile antennas

来源：评论

学校读者我要写书评

暂无评论

Design Objectives for Synthesis of Graphene PN Junction Circuits Based on Two-Level Representation

Design Objectives for Synthesis of Graphene PN Junction Circ...

引用

Euromicro Symposium on Digital System Design

作者： Subrata Das Arighna Deb Petr Fiser Debesh Kumar Das Dept of Computer Science & Engineering Supreme Knowledge Foundation Group of Institutions Hooghly India School of Electronics Engineering Kalinga Institute of Industrial Technology Bhubaneswar India Faculty of Information Technology Czech Technical University Prague Czech Republic Dept. of Computer Science & Engineering Jadavpur University Kolkata India

ISBN: (数字)9798350380385

ISBN: (纸本)9798350380392

The development of electrostatically doped graphene PN-junctions shows promise for creating efficient low-power, high-speed circuits. In recent years, there has been a considerable interest in the synthesis of graphene PN junction logic circuits. However, existing synthesis methods lack assessment based on technology-specific cost metrics (e.g., the number of graphene PN junction gates, constant inputs), leading to insufficiently addressed design objectives. In this paper, we introduce synthesis approaches for graphene PN-junction circuits based on Sum-of-Products (SoP) and Exclusive Sum-of-Products (ESoP) function representations. Experimental results indicate that ESoP-based synthesis significantly reduces the number of graphene PN junction gates, constant inputs, and switching activity compared to SoP-based approaches. Overall, ESoP-based synthesis is deemed more suitable than SoP-based methods for designing graphene PN-junction logic circuits.

关键词： Measurement Costs Digital systems Logic circuits Graphene Switches Logic gates Circuit synthesis Logic Junctions

来源：评论

学校读者我要写书评

暂无评论

Cloud-based Machine Learning Adoption Model for Higher Education Institutions

Cloud-based Machine Learning Adoption Model for Higher Educa...

引用

2024 International Conference on Artificial Intelligence and Emerging technology, Global AI Summit 2024

作者： Kiros, Atakilti Brhanu Salau, Ayodeji Olalekan Vats, Satvik Kaushik, Keshav Tin, Ting Tin Omeje, Crescent Onyebuchi Mekelle University Mekelle Institute of Technology Mit Mekelle Ethiopia Afe Babalola University Department of Electrical/Electronics and Computer Engineering Ado-Ekiti Nigeria Department of Cse Graphic Era Hill University Uttarakhand Dehradun India Graphic Era Deemed to be University Uttarakhand Dehradun248007 India Amity School of Engineering and Technology Amity University Mohali Punjab India Inti International University Faculty of Data Science and Information Technology Negeri Sembilan Malaysia University of Port Harcourt Department of Electrical/Electronic Engineering Rivers State Choba500004 Nigeria

ISBN: (纸本)9798350379716

As a new concept, cloud computing has attracted the IT Initiations attention especially the Mobile learning (M-learning). At present, there Higher Education Institutions (HEI's) have struggle with limited IT budgets, difficulties in obtaining and managing infrastructure licenses, and the complexities of software and hardware management. The coming of cloud computing, all of the above challenges was solved. The migration to cloud based M-learning in HEI's is a significant stage in terms of online education, especially during the COVID-19 pandemic. Cloud based M-learning approach has the potential to be a powerful and rapid solution to the many challenges faced by universities in this unprecedented time. This study starts with some brief concepts of Cloud Computing and M-learning for HEI's, mentioning for the significance of the results obtained so far. Additionally, it starting point as strategy for HEI's to adopt Cloud based M-learning is provided, by comparisons of existing models and select DOI model to propose cloud based adoption model. But, the DOI model has its own weaknesses;for example, it does not take into account the localization as Innovation determinants. This study introduces localization phase for the optimal performance of the adopted model. The adoption model was evaluating the importance of the proposed model by IT domain experts and it initiates to be both useful and suitable for a successful cloud based M-learning adoption Ethiopian perspective. Therefore, the research results obtained are promising and funding the use of Cloud based M-learning solutions for Ethiopia HEI's by improving knowledge and skill in this field provides a useful strategy for the HEI's structure. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

A cost- and energy-efficient SRAM design based on a new 5 i-p majority gate in QCA nanotechnology

引用

Materials Science and engineering: B 2024年 302卷

作者： Kassa, Sankit Ahmadpour, Seyed-Sajad Lamba, Vijay Kumar Misra, Neeraj Jafari Navimipour, Nima Kotecha, Ketan Maharashtra Pune India School of Electronics Engineering VIT-AP University Andhra Pradesh Amaravathi522237 India Department of Computer Engineering Faculty of Engineering and Natural Sciences Kadir Has University Istanbul Turkey Future Technology Research Center National Yunlin University of Science and Technology Douliou64002 Taiwan

Quantum-dot Cellular Automata (QCA) is a revolutionary paradigm in the Nano-scale VLSI market with the potential to replace the traditional Complementary Metal Oxide Semiconductor system. To demonstrate its usefulness, this article provides a QCA-based innovation structure comprising a 5-input (i-p) Majority Gate, which is one of the basic gates in QCA, and a Static Random Access Memory (SRAM) cell with set and reset functionalities. The suggested design, with nominal clock zones, provides a reliable, compact, efficient, and durable configuration that helps achieve the optimal size and latency while decreasing power consumption. Based on the suggested 5 i-p majority gate, the realized SRAM architecture improves energy dissipation by 33.95 %, cell count by 31.34 %, and area by 33.33 % when compared to the most recent design designs. Both the time and the cost have been decreased by 30 % and 53.95 %, respectively. © 2024

关键词： Energy dissipation

来源：评论

学校读者我要写书评

暂无评论

A Minimized Thickness of MIMO Antenna Based on Thin Film Substrates for UWB application

A Minimized Thickness of MIMO Antenna Based on Thin Film Sub...

引用

International Electrical engineering Congress (iEECON)

作者： Pongsathorn Aroonmitr Chawalit Rakluea Wisit Loedhammacakra Paitoon Rakluea Maleeya Tangjitjetsada Chatree Mahatthanajatuphat Prayoot Akkaraekthalin Tawan Thianthong Department of Electronics and Telecommunication Engineering Faculty of Engineering Rajamangala University of Technology Thanyaburi Pathum Thani Thailand Department of Electronics and Telecommunication Engineering Faculty of Engineering and Architecture Rajamangala University of Technology Suvarnabhumi Nonthaburi Thailand Department of Electrical and Computer Engineering Faculty of Engineering King Mongkut's University of Technology North Bangkok Bangkok Thailand Siam Airport Ground Services Co. Ltd. Bangkok Thailand

This paper presents a compact MIMO antenna for ultra-wideband (UWB) applications. It is created based on a thin film for an antenna substrate to minimize the antenna thickness, resulting in the antenna being utilized well on a curved surface. The film thickness is about 0.056 mm was used to design and cooperate with the rectangular slot antenna and rectangular stub. In addition, the antenna cover frequency range from 3.1 GHz to 10.6 GHz with the S11 and S21 below −10 dB and −15 dB, respectively. The antenna will be fed by $50\ \Omega$ coplanar waveguide for a transmission line. Moreover, an Envelope Correlation Coefficient (ECC) isn't more than 0.05 throughout the frequency range. The antenna radiation pattern is bidirectional all bands and the maximum gain is 4.6 dBi. The proposed antenna can be applied to the WBAN WPAN and WLAN, UWB communication systems as well.

关键词： Correlation coefficient Wireless communication Wireless LAN Slot antennas Wireless personal area networks Transmission line measurements Ultra wideband antennas

来源：评论

学校读者我要写书评

暂无评论

Utilizing Advanced Techniques for the Extraction of Handwritten Text in Challenging Scenarios and Diverse Languages

Utilizing Advanced Techniques for the Extraction of Handwrit...

引用

2023 International Conference for Technological engineering and its Applications in Sustainable Development, ICTEASD 2023

作者： Singh, Devendra Rathika, S. Al-Fatlawy, Ramy Riad Kumar Tiwari, Anurag Sapaev, B. Chaturvedi, Rishabh Chythanya, Nagaraju Krishna Uttaranchal Institute of Technology Uttaranchal University Department of Computer Science & Engineering Dehradun248007 India MBA Prince Shri Venkateshwara Padmavathy Engineering College Chennai127 India The Islamic University Najaf Iraq IES College of Technology Department of Electronics & Communication Engineering Madhya Pradesh Bhopal462044 India Faculty of Medicine Alfraganus University Department of Pharmaceuticals and Chemistry Tashkent Uzbekistan GLA University Department of Mechanical Engineering Mathura India GRIET Telangana Hyderabad India

ISBN: (纸本)9798350336474

Handwritten text recognition is the computer's ability to capture and interpret handwritten input from various sources, including paper documents, touchscreen interfaces, photo scans, and more. It falls within the domain of pattern recognition, [16] specifically character recognition, aiming to transform spatially represented handwritten language into its symbolic representation. Each script consists of characters or letters with distinctive basic shapes. The primary objective of handwriting recognition is to accurately identify the input characters or images, which are subsequently processed by automated systems. This technology finds applications in recognizing various forms of handwritten content, encompassing digits, numerals, cursive text, symbols, and scripts in multiple languages, including English. The automated recognition of handwritten text holds significant utility across numerous applications where handling substantial volumes of handwritten data is crucial. These applications include address and postcode recognition on envelopes, amounts on bank checks, document analysis, and signature verification. Consequently, computers play a pivotal role in facilitating document processing, whereby handwritten text is scanned, converted into editable documents, and can be translated into local languages like Hindi for increased accessibility and usability. © 2023 IEEE.

关键词： Optimization

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：