检索结果-内蒙古大学图书馆

CRD-CGAN:category-consistent and relativistic constraints for diverse text-to-image generation

Frontiers of computer Science 2024年第1期18卷 61-75页

作者： Tao HU Chengjiang LONG Chunxia XIAO College of Intelligent Systems Science and Engineering Hubei Minzu UniversityEnshi 445000China School of Computer Science Wuhan UniversityWuhan 430072China Key Laboratory of Performing Art Equipment&System Technology Ministry of Culture and TourismBeijing 100007China Meta Reality Labs BurlingameCA94010USA

Generating photo-realistic images from a text description is a challenging problem in computer *** works have shown promising performance to generate synthetic images conditional on text by Generative Adversarial Networks(GANs).In this paper,we focus on the category-consistent and relativistic diverse constraints to optimize the diversity of synthetic *** on those constraints,a category-consistent and relativistic diverse conditional GAN(CRD-CGAN)is proposed to synthesize K photo-realistic images *** use the attention loss and diversity loss to improve the sensitivity of the GAN to word attention and ***,we employ the relativistic conditional loss to estimate the probability of relatively real or fake for synthetic images,which can improve the performance of basic conditional ***,we introduce a category-consistent loss to alleviate the over-category issues between K synthetic *** evaluate our approach using the Caltech-UCSD Birds-200-2011,Oxford 102 flower and MS COCO 2014 datasets,and the extensive experiments demonstrate superiority of the proposed method in comparison with state-of-the-art methods in terms of photorealistic and diversity of the generated synthetic images.

关键词： text-to-image diverse conditional GAN relativi-stic category-consistent

来源：评论

学校读者我要写书评

暂无评论

Improved LOPCOW-SAW Method for Optimal Supplier Selection in Supply Chain Management 12

Improved LOPCOW-SAW Method for Optimal Supplier Selection in...

引用

12th International Conference on Cyber and IT Service Management, CITSM 2024

作者： Sumanto Sumarna Nurdin, Hafis Asra, Taufik Sobari, Irwan Agus Kuswanto, Herman Chaidir, Indra Akil, Ibnu Handono, Felix Wuryo Universitas Bina Sarana Informatika Informatics Faculty of Engineering and Informatics Jakarta Indonesia Universitas Nusa Mandiri Information System Faculty of Engineering and Informatics Jakarta Indonesia Universitas Bina Sarana Informatika Software Engineering Faculty of Engineering and Informatics Jakarta Indonesia Universitas Nusa Mandiri Universitas Bina Sarana Informatika Information System Faculty of Engineering and Informatics Indonesia Universitas Nusa Mandiri Universitas Bina Sarana Informatika Computer Engineering Faculty of Engineering and Informatics Indonesia

ISBN: (纸本)9798331528799

Supplier selection is a critical component of Supply Chain Management (SCM), as suppliers provide raw materials or services essential for supporting a company's operations. Traditional supplier selection methods often rely on subjective assessments to determine the weighting of criteria, which can lead to biases and suboptimal decision making. To address this issue, this study combines the Logarithmic Percentage Change of Weight (LOPCOW) method with the Simple Additive Weighting (SAW) method to enhance objectivity in the supplier selection process. The LOPCOW method is used to assign criterion weights objectively based on logarithmic percentage changes, which helps minimize subjectivity, These weights are then applied using the SAW method to evaluate and rank suppliers. The results of the study show that this combined approach is more reliable for selecting suppliers objectively and accurately. The study suggests integrating other Multi-Criteria Decision-Making (MCDM) methods to further strengthen the results while recommending the inclusion of real-time data analytics to improve the model's adaptability to dynamic market conditions. The LOPCOW-SAW approach can be applied across various industries for more effective and efficient SCM supplier selection. © 2024 IEEE.

关键词： Supply chain management

来源：评论

学校读者我要写书评

暂无评论

Adaptive Score Normalization using Machine Learning for enhanced Fingerprint Recognition 2

Adaptive Score Normalization using Machine Learning for enha...

引用

2nd International Conference on Self Sustainable Artificial Intelligence systems, ICSSAS 2024

作者： Bhargava, Neeraj Rathore, Pramod Singh Goswami, Ankur Bhowmick, Apoorva Singh, Jatin Mds University School of Engineering & System Sciences Ajmer India Manipal University Department of Computer and Communication Engineering Jaipur India Govt. Engineering College Department of Computer Applications Bikaner India United Nations Secretariate New YourkNY10017 United States Baghdad Iraq Mds University Ajmer India

ISBN: (纸本)9798350368413

Fingerprint identification systems play a vital role in biometric security but are often influenced by fluctuations in scanning settings and environmental variables. The conventional normalizing techniques such as Min-Max, Z-Score, and Tanh normalization provide fixed solutions that do not adjust to these variations, therefore degrading the accuracy and resilience of the results. This study presents a machine learning-based adaptive normalization system that seamlessly adapts to present circumstances by integrating contextual information, such as ambient variables and scanner-specific features, with raw fingerprint scores. The procedure includes the gathering of data, doing preprocessing, training models using decision trees, random forests, and neural networks, and implementing dynamic normalization. Comprehensive testing conducted on several scanners and settings has shown the efficacy of this method, demonstrating substantial improvements in both recognition accuracy and resilience. Assessment measures such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and R2 Score validate the higher efficacy of this adaptive approach in comparison to static methods, thereby improving operational adaptability and establishing a basis for wider biometric identification applications. © 2024 IEEE.

关键词： Random forests

来源：评论

学校读者我要写书评

暂无评论

Backup subscription scheme for differential protection using IEC61850-9-2 sampled values

引用

Energy Informatics 2024年第1期7卷 110页

作者： Katoulaei, Mohammad Khalili Rahmani, Aamir Høidalen, Hans Kristian Oleinikova, Irina Mork, Bruce Department of energy systems Norwegian University of Science and Technology Trondheim Norway Department of electrical power system Michigan Technical University Michigan United States Department of Electrical Power Engineering Norwegian University of Science and Technology Trondheim Norway Department of Electrical and Computer Engineering Michigan Technological University HoughtonMI49931 United States

In IEC-61850-based digital substations, the protection IED’s performance is dependent on merging unit’s vendor implementation, communication networks, and measurement circuit’s health conditions. As the process bus Sampled Value(SV) enables the availability of all sensor data on a communication network, this paper proposes a Backup Subscription scheme (BSS) for a transformer differential protection (87T, PDIF) function. BSS utilizes sensor data in digital substations to achieve a flexible protection scheme using a dynamic subscription feature. Thus, in case of failure of one sensor, differential protection would be maintained. The paper presents the implementation and verification of a prototyped scheme using a Hardware-in-the-loop simulation test bed. The main result is that BSS integration into differential protection ensures its dependability and security. Moreover, delay compensation and seamless switching feature increases the availability of differential protection. © The Author(s) 2024.

关键词： Hardware-in-the-loop simulation

来源：评论

学校读者我要写书评

暂无评论

A Frequency Response Based Methodology for the Design and Analysis of Active Power Filters Current Control

A Frequency Response Based Methodology for the Design and An...

引用

2024 IEEE Energy Conversion Congress and Exposition, ECCE 2024

作者： Sánchez, Lidia Reigosa, David Sola, Javier Agorreta, Juan Luis Balda, Julian Briz, Fernando University of Oviedo Dept. of Elect. Computer & System Engineering Gijón33204 Spain Ingeteam Power Technology Photovoltaic Solar Energy Sarriguren31621 Spain

ISBN: (纸本)9798350376067

Design and tuning of active power filters (APFs) current control is not trivial due to the multiple control objectives (e.g., reactive power and/or harmonic compensation) and options for the controllers' design (i.e., combinations of proportional-resonant and/or proportional-integral current regulators in different reference frames). Given the large amount of possible control configuration, the use of numerical simulations can be impractical. This paper proposes a methodology using analytical complex vector models combined with frequency response functions, aimed to systematize the performance analysis of APF current control. The proposed approach will enable the comparison of an arbitrarily large number of control configurations and tunings based on predefined performance indices without the use of time simulations. © 2024 IEEE.

关键词： Active filters

来源：评论

学校读者我要写书评

暂无评论

Towards Stricter Black-box Integrity Verification of Deep Neural Network Models 24

Towards Stricter Black-box Integrity Verification of Deep Ne...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： He, Chaoxiang Bai, Xiaofan Ma, Xiaojing Zhu, Bin B. Hu, Pingyi Fu, Jiayun Jin, Hai Zhang, Dongmei Huazhong University of Science and Technology Hubei Wuhan China Shanghai Jiao Tong University Shanghai China Microsoft Corporation Beijing China National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Engineering Research Center on Big Data Security Hubei Key Laboratory of Distributed System Security School of Cyber Science and Engineering Huazhong University of Science and Technology China National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology China

ISBN: (纸本)9798400706868

Cloud-based machine learning services offer significant advantages but also introduce the risk of tampering with cloud-deployed deep neural network (DNN) models. Black-box integrity verification (BIV) allows model owners and end-users to determine if a cloud-deployed DNN model has been tampered with by examining only the top-1 label responses. Fingerprinting generates fingerprint samples to query the model, achieving BIV with no impact on the model's accuracy. In this paper, we present BIVBench, the first comprehensive benchmark for BIV of DNN models. BIVBench covers 16 types of model modifications, providing extensive coverage of practical modification scenarios. Our analysis reveals that existing fingerprinting methods, which are typically focused on significant tampering, lack the sensitivity needed to effectively detect subtle yet common and potentially severe modifications. To address this limitation, we propose MiSentry (Model Integrity Sentry), a novel fingerprinting method that leverages meta-learning. MiSentry strategically incorporates a few subtly modified models into the meta-learning model zoo and maximizes the divergence of output predictions between the target model and the modified models in the model zoo to generate highly sensitive, generalizable, and effective fingerprint samples. Extensive evaluations using BIVBench demonstrate that MiSentry outperforms existing state-of-the-art methods overall and significantly surpasses them in detecting subtle modifications. The BIVBench and supplementary materials are available at: https://***/CGCL-codes/BIVBench. © 2024 ACM.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

system for Talent Acquisition: Integrating AI, Automation, and Data Analysis in HR 6

System for Talent Acquisition: Integrating AI, Automation, a...

引用

6th International Conference on Artificial Intelligence in Information and Communication, ICAIIC 2024

作者： Abitova, Gulnara Serikov, Ayanbek Nikulin, Vladimir Rakhimzhanova, Mira Shuteyeva, Gulnur Kulniyazova, Korlan Astana IT University Dep. Computer Engineeering Astana Kazakhstan Binghamton University Dept. Electical Engineering and Computer Science Binghamton United States Eurasian National University Dept. System Anslysis and Managemnt Astana Kazakhstan

ISBN: (纸本)9798350344349

This article presents a novel system designed for human resources (HR) companies, aiming to predict an individual's personality based on resume analysis through the integration of advanced information technology tools. At the core of this system is Airable, serving as the central database, which is seamlessly I integrated with various other tools such as *** for document handling, GitHub for code repository management, and OpenAI for advanced AI algorithms. The integration and automation of these diverse tools are skillfully orchestrated using Make, a no-code solution that streamlines the workflow and enhances efficiency. This innovative approach allows HR companies to leverage technology for more accurate personality predictions, thereby improving their recruitment processes and decision-making strategies. © 2024 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Modeling, Analysis and Reduction of Radiated EMI Due to the Pulsating Voltage on PCB Ground Plane in a Non-Isolated Power Converter

Modeling, Analysis and Reduction of Radiated EMI Due to the ...

引用

2024 IEEE Energy Conversion Congress and Exposition, ECCE 2024

作者： Lai, Yanwen Wang, Shuo Yang, Yirui Huang, Qinghui Luo, Zheng University of Florida Department of Electrical and Computer Engineering GainesvilleFL United States University of Florida Monolithic Power System Inc. San JoseCA United States

ISBN: (纸本)9798350376067

Electromagnetic interference (EMI) from power converters is a critical concern, subject to rigorous standards across multiple sectors. The transition to high-frequency switching and the increasing use of wide bandgap semiconductors have elevated radiated EMI to a primary design consideration. To address these issues, existing research in EMI modeling and reduction techniques is essential for developing power conversion applications that meet rigorous compliance requirements. The radiated EMI mechanisms of power cable antennas and components were studied and modeled, however, other factors could contribute to radiation when the impact of power cables antenna and components on radiated EMI could be neglected. In this paper, the radiated EMI due to the pulsating voltage between the converter and earth ground is investigated. Based on the converter printed circuit board (PCB) layout and test setup, the radiation model is proposed, analyzed and verified in experiments. © 2024 IEEE.

关键词： Antenna grounds

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Approach with BRKGA and Data Mining for the Early/Tardy Scheduling Problem 13

A Hybrid Approach with BRKGA and Data Mining for the Early/T...

引用

13th IEEE Congress on Evolutionary Computation, CEC 2024

作者： Mendonça, Israel Fatyanosa, Tirana Noor Aritsugi, Masayoshi Silva, Pedro Henrique González Kumamoto University Faculty of Advanced Science and Technology Kumamoto Japan Brawijaya University Faculty of Computer Science Malang Indonesia Federal University of Rio de Janeiro System Engineering and Computer Science Program Rio de Janeiro Brazil

ISBN: (纸本)9798350308365

This paper introduces a novel hybrid genetic algorithm combined with data mining to solve a version of the early/tardy scheduling problem in which no unforced idle time may be inserted in a sequence. The chromosome representation of the problem is based on random keys and we use the Biased Random-Key Genetic Algorithm (BRKGA) to establish an order in which jobs are scheduled. A data mining component gathers data from the evolutionary steps of BRKGA and suggests chromosomes based on past observations. In this way, we address a key challenge in BRKGA - the exploration of solutions near an individual gene - by leveraging data-driven insights to refine and enhance the search space. Comparative analysis reveals that our hybrid algorithm significantly benefits from the pattern recognition capabilities of data mining, leading to improved scheduled efficiency. The results of a series of computational experiments underscore the potential of this hybrid approach that, although requiring slightly longer computational times, is better than the previous baseline BRKGA algorithm in terms of solution quality. © 2024 IEEE.

关键词： Scheduling algorithms

来源：评论

学校读者我要写书评

暂无评论

A Novel Approach for Achieving FOMs in Space Constrained Optimized Arrays for AESA Applications

A Novel Approach for Achieving FOMs in Space Constrained Opt...

引用

2024 International Conference on Microelectronics, ICM 2024

作者： Naveed, Athar Tahir, Ghiayas Hassan, Arshad Ali, Shawkat Khan, Arshad Bermak, Amine Nastp Electronics System Design Center Pakistan National University of Computer and Emerging Sciences Department of Electrical Engineering Islamabad Pakistan College of Science and Engineering Hamad Bin Khalifa University Qatar Foundation Doha Qatar

ISBN: (纸本)9798350379396

This paper defines an approach applied for implementation of application specific beamforming modalities in finite aperture, space-constrained, optimized arrays. To achieve desired response in array factor, appropriate tapering schemes are selected while focusing on array Figures of Merits (FOMs). In this paper, space-constrained optimized arrays of sizes 1 × 32 (linear) and 8 × 8 (planar) are designed using Opti-metrics in High Frequency Structure Simulator (HFSS). The calculated weights and steering scheme are then applied to HFSS models and Matlab based calculations are validated for proof of concept. Array metrics like low SLL, limited beam-broadening, increased directivity and limited scan loss in planned Field of View (FOV) are targeted metrics. This paper also briefly elaborates comparison of various tapering techniques, while focusing on defined FOMs. The working procedure is developed which extends itself from excitation control of linear to planar dimensions. Keeping in view limited computational resources for full wave simulation, an approach using Pattern Multiplication (PM) is suggested to converge to desired results for large sized arrays. The scheme of this paper has also been used in a design of Hybrid beamforming network (BFN). © 2024 IEEE.

关键词： MATLAB

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：