检索结果-内蒙古大学图书馆

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

作者： Zhang, Ao Wang, He Guo, Pengcheng Fu, Yihui Xie, Lei Gao, Yingying Zhang, Shilei Feng, Junlan School of Computer Science Xi'an China China Mobile Research Institute Beijing China

ISBN: (纸本)9781728163277

The performance of the keyword spotting (KWS) system based on audio modality, commonly measured in false alarms and false rejects, degrades significantly under the far field and noisy conditions. Therefore, audio-visual keyword spotting, which leverages complementary relationships over multiple modalities, has recently gained much attention. However, current studies mainly focus on combining the exclusively learned representations of different modalities, instead of exploring the modal relationships during each respective modeling. In this paper, we propose a novel visual modality enhanced end-to-end KWS framework (VE-KWS), which fuses audio and visual modalities from two aspects. The first one is utilizing the speaker location information obtained from the lip region in videos to assist the training of multi-channel audio beamformer. By involving the beamformer as an audio enhancement module, the acoustic distortions, caused by the far field or noisy environments, could be significantly suppressed. The other one is conducting cross-attention between different modalities to capture the inter-modal relationships and help the representation learning of each modality. Experiments on the MSIP challenge corpus show that our proposed model achieves a 2.79% false rejection rate and a 2.95% false alarm rate on the Eval set, resulting in a new SOTA performance compared with the top-ranking systems in the ICASSP2022 MISP challenge. © 2023 IEEE.

关键词： Alarm systems

来源：评论

学校读者我要写书评

暂无评论

A 2 Kbits Low Power EEPROM for Passive RFID Tag IC

引用

Chinese Journal of Electronics 2022年第1期31卷 18-24页

作者： HU Jianguo WANG Deming WU Jing School of Microelectronics Science and Technology Sun Yat-sen University School of Physics and Telecommunication Engineering South China Normal University Development Research Institute of Guangzhou Smart City

This paper presents a low power consumption and low cost electrically erasable programmable read-only memory(EEPROM) for radio frequency identification(RFID) tag chip. A read-write circuit with parallel input and serial output is proposed. Only one sensitive amplifier is used to read the data in memory,which can effectively reduce the power consumption of the read operation. Because the tag may be read or written while moving, the internal voltage may change in a wide range. Therefore, this paper designs a charge pump and its control circuit with wide voltage working *** proposed EEPROM is integrated into an RFID tag chip and fabricated using a 180 nm complementary metal-oxide semiconductor(CMOS) process. Experimental results show that the circuit can work in the input voltage range of 1 V to 1.8 V, and the minimum current of read operation and write operation is 0.68 μA and 30 μA respectively, which has the characteristic of low power consumption.

关键词： wide voltage working range low power EEPROM EPROM control circuit serial output input voltage range radio frequency identification tag chip complementary metal-oxide semiconductor process passive RFID tag IC read-write circuit low-power electronics internal voltage low cost electrically erasable programmable read-only memory parallel input CMOS integrated circuits radiofrequency identification RFID tag chip sensitive amplifier low power consumption read operation CMOS process

来源：评论

学校读者我要写书评

暂无评论

Reviewing metal supported solid oxide fuel cells for efficient electricity generation with biofuels for mobility

引用

Journal of Energy Chemistry 2025年第4期103卷 106-153页

作者： Fábio C.Antunes João P.J.de Oliveira Ricardo S.de Abreu Thiago Dias Bruno B.N.S.Brandão Josué M.Gonçalves Josimar Ribeiro Julian Hunt Hudson Zanin Gustavo Doubek School of Electrical and Computer Engineering University of CampinasCampinas 13083-852Brazil School of Food Engineering University of CampinasCampinas 13083-852Brazil School of Chemical Engineering University of CampinasCampinas 13083-852Brazil Mackenzie Institute for Research in Graphene and Nanotechnologies(MackGraphe) Mackenzie Presbyterian InstituteSão Paulo 01302-907Brazil Electrochemistry Research and Development Laboratory(LPDE) Department of ChemistryCenter of Exact SciencesFederal University of Espírito SantoCampus GoiabeirasVitória 29075-910Brazil Biological and Environmental Science and Engineering Division King Abdullah University of Science and TechnologyThuwal 23955-6900Saudi Arabia

Metal-Supported Solid Oxide Fuel Cells(MS-SOFCs)hold significant potential for driving the energy *** electrochemical devices represent the most advanced generation of Solid Oxide Fuel Cell(SOFCs)and can pave the way for mass production and wider adoption than Proton Exchange Membrane Fuel Cells(PEMFCs)due to their fuel flexibility,higher power density and the absence of noble metals in the fabrication *** review examines the state-of-the-art of SOFCs and MS-SOFCs,presenting perspectives and research directions for these key technological devices,highlighting novel materials,techniques,architectures,devices,and degradation mechanisms to address current challenges and future *** such as infiltration/impregnation,ex-solution catalyst synthesis,and the use of a pre-catalytic reformer layer are discussed as their impact on efficiency and prolonged *** concepts are also described and connected with well-dispersed nano particles,hindrance of coarsening,and an increased number of Triple Phase Boundaries(TPBs).This review also describes the synergistic use of reformers with MS-SOFCs to compose solutions in energy generation from readily available ***,the End-of-Life(EoL),recycling,and life-cycle assessments(LCAs)of the Fuel Cell Hybrid Electric Vehicles(FCHEVs)were *** comparing Fuel Cell Electric Vehicles(FCEVs)equipped with(PEMFCs)and FCHEVs equipped with MS-SOFCs,both powered with hydrogen(H_(2))generated by different routes were *** review aims to provide valuable insights into these key technological devices,emphasizing the importance of robust research and development to enhance performance and lifespan while reducing costs and environmental impact.

关键词： Reform Metal-supported solid oxide fuel cell Powertrain systems Fuel cell Hybrid electric vehicle Biofuel

来源：评论

学校读者我要写书评

暂无评论

HHO-ACO hybridized load balancing technique in cloud computing

引用

International Journal of Information Technology (Singapore) 2023年第3期15卷 1357-1365页

作者： Sumathi, Muruganandam Vijayaraj, Natarajan Raja, Soosaimarian Peter Rajkamal, Murugesan School of Computing SASTRA Deemed University Tamil Nadu Thanjavur India Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala Research and Development Institute of Science and Technology Tamil Nadu Chennai India School of Computer Science and Engineering Vellore Institute of Technology Tamil Nadu Vellore India IBM Karnataka Bangalore India

Due to on-demand requirement of computing resources, cloud computing (CC) is the widely used technology in data storage, software and platform. CC provides everything as a service to requester. In an Infrastructure as... 详细信息

Due to on-demand requirement of computing resources, cloud computing (C_C) is the widely used technology in data storage, software and platform. C_C provides everything as a service to requester. In an Infrastructure as a Service (IaaS), virtual machines (VMs) play a vital role to provide infrastructure to the requester. The performance VMs depends on the distribution of work (load balancing) between VMs. The process of distributing a set of tasks or workloads over a set of VM resources is called as Load Balancing (LB). Unbalanced task distribution leads to overloaded or under-loaded issues and performance degradation. Hence, LB is a desirable and essential task in C_C. To improve the LB performance the two different meta-heuristic optimization algorithms, Harries Hawks Optimization (HHO) and Ant Colony Optimization (ACO) are hybridized in the proposed technique. The hybridized load balancing (HLB) algorithm performance is compared to HHO and ACO. The factors that are analyzed in the proposed system are the average waiting time (A_WT), average execution time (A_ET), average response time (A_RT), make-span, throughput analysis, turnaround time and LB time. The HLB mechanism is implemented and simulated in java and Cloudsim respectively. By the allocation of workloads to VMs will shows which one gives the best efficiency in LB among the VM’s. The proposed HLB technique takes 9.29% A_ET, 2.65% of A_RT, 0% of A_WT, less turnaround time and LB time than the basic HHO and ACO algorithms. Thus, the proposed HLB provided better performance than HHO and ACO is proven with different performance metrics. © 2023, The Author(s), under exclusive licence to Bharati Vidyapeeth's institute of computer Applications and Management.

关键词： Ant colony optimization Cloud computing Harries haws optimization Hybridized load balancing Virtual machine

来源：评论

学校读者我要写书评

暂无评论

Sample Complexity of Posted Pricing for a Single Item 38

Sample Complexity of Posted Pricing for a Single Item

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Jin, Billy Kesselheim, Thomas Ma, Will Singla, Sahil Cornell University School of Operations Research and Information Engineering United States University of Bonn Institute of Computer Science Lamarr Institute for Machine Learning and Artificial Intelligence Germany Columbia University Graduate School of Business Data Science Institute United States Georgia Tech School of Computer Science United States

Selling a single item to n self-interested buyers is a fundamental problem in economics, where the two objectives typically considered are welfare maximization and revenue maximization. Since the optimal mechanisms are often impractical and do not work for sequential buyers, posted pricing mechanisms, where fixed prices are set for the item for different buyers, have emerged as a practical and effective alternative. This paper investigates how many samples are needed from buyers' value distributions to find near-optimal posted prices, considering both independent and correlated buyer distributions, and welfare versus revenue maximization. We obtain matching upper and lower bounds (up to logarithmic factors) on the sample complexity for all these settings. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An integrated machine learning model for accurate and robust prediction of superconducting critical temperature

引用

Journal of Energy Chemistry 2023年第3期78卷 232-239,I0007页

作者： Jingzi Zhang Ke Zhang Shaomeng Xu Yi Li Chengquan Zhong Mengkun Zhao Hua-Jun Qiu Mingyang Qin X.-D.Xiang Kailong Hu Xi Lin School of Materials Science and Engineering Harbin Institute of TechnologyShenzhen 518055GuangdongChina Blockchain Development and Research Institute Harbin Institute of TechnologyShenzhen 518055GuangdongChina State Key Laboratory of Advanced Welding and Joining Harbin Institute of TechnologyHarbin 150001HeilongjiangChina School of Materials Science and Engineering Harbin Institute of TechnologyHarbin 150001HeilongjiangChina Department of Materials Science and Engineering&Department of Physics Southern University of Science and TechnologyShenzhen 518055GuangdongChina

Discovering new superconductors via traditional trial-and-error experimental approaches is apparently a time-consuming process,and the correlations between the critical temperature(Tc) and material features are still *** rise of machine learning(ML) technology provides new opportunities to speed up inefficient exploration processes,and could potentially uncover new hints on the unclear *** this work,we utilize open-source materials data,ML models,and data mining methods to explore the correlation between the chemical features and Tcvalues of superconducting *** further improve the prediction accuracy,a new model is created by integrating three basic algorithms,showing an enhanced accuracy with the coefficient of determination(R2) score of 95.9 % and root mean square error(RMSE) of 6.3 *** average marginal contributions of material features towards Tcvalues are estimated to determine the importance of various features during prediction *** results suggest that the range thermal conductivity plays a critical role in Tcprediction among all element ***,the integrated ML model is utilized to screen out potential twenty superconducting materials with Tcvalues beyond 50.0 *** study provides insights towards Tcprediction to accelerate the exploration of potential high-Tcsuperconductors.

关键词： Superconductors Integrated machine learning Superconducting critical temperature

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of RESNET-Pretrained Models for computer Vision 2023

A Comparative Study of RESNET-Pretrained Models for Computer...

引用

15th International Conference on Contemporary Computing, IC3 2023

作者： Prajwal, Thode Sai Ilavarasi, A.K. School of Computer Science and Engineering Vellore Institute of Technology-Chennai India School of Computer Science and Engineering Centre for Healthcare Advancement Innovation and Research Vellore Institute of Technology-Chennai India

ISBN: (纸本)9798400700224

The science of computer vision has recently seen a revolution with deep learning models, and Residual Networks (ResNets) have become one of the most popular and efficient designs. ResNets have outperformed conventional convolutional neural networks in several image classification tasks, demonstrating outstanding results. An overview of ResNets and their variants, including Res2Net, COVID-ResNet, ResNeSt, SENet, and others, is presented in this review. We assess the performance of the ResNet design variants using benchmark datasets, highlight the major ideas and technological advancements underlying them, and their pros and cons. We also discuss some recent developments in ResNets research and applications, including object identification, picture segmentation, and transfer learning. This review would be beneficial to the research community in selecting the appropriate models based on the intended usage. © 2023 ACM.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

Enhancing Healthcare Record Privacy Through the Integration of Visual Cryptography and Diverse Image Encryption Techniques 9

Enhancing Healthcare Record Privacy Through the Integration ...

引用

9th International Conference on Communication and Electronics Systems, ICCES 2024

作者： Chinnasamy, P. Deepalakshmi, P. Sandeep, Dayina Ganesh, Addala Siva Krishna, Angalakurthi Jaya Priya, D. Divya School of Computing Kalasalingam Academy of Research and Education Department of Computer Science and Engineering Tamilnadu India Kalasalingam Academy of Research and Education Department of Computer Science and Engineering Krishnankoil Virudhunagar India Mlr Institute of Technology Department of Computer Science and Engineering Hyderabad India

ISBN: (纸本)9798350377972

The methodology of this research focuses on analyzing real-world breaches, such as the Star Health Insurance leak, and demonstrating how multi-layered encryption techniques could have mitigated the impact. By splitting sensitive healthcare information into multiple shares using visual cryptography, unauthorized access is prevented unless all shares are combined, adding an additional layer of security. Furthermore, image-based encryption adds another line of defense, ensuring that even if hackers gain access to data, it remains unusable without the correct decryption methods. Through the implementation of these techniques, we demonstrate a significant enhancement in the security of healthcare records. By combining diverse encryption methods, our approach not only strengthens the protection of personal health information but also prevents unauthorized data dissemination, as seen in the Star Health breach. The results indicate that our model is resilient to various cyber-attacks, reducing the likelihood of large-scale data leaks and improving the integrity of healthcare systems. © 2024 IEEE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

EID: Facilitating Explainable AI Design Discussions in Team-Based Settings

引用

International Journal of Crowd science 2023年第2期7卷 47-54页

作者： Zhang, Jiehuang Yu, Han School of Computer Science and Engineering Nanyang Technological University Singapore639798 Singapore Alibaba-NTU Singapore Joint Research Institute Singapore637335 Singapore

Artificial intelligence (AI) systems have many applications with tremendous current and future value to human society. As AI systems penetrate the aspects of everyday life, a pressing need arises to explain their decision-making processes to build trust and familiarity among end users. In high-stakes fields such as healthcare and self-driving cars, AI systems are required to have a minimum standard for accuracy and to provide well-designed explanations for their output, especially when they impact human life. Although many techniques have been developed to make algorithms explainable in human terms, no design methodologies that will allow software teams to systematically draw out and address explainability-related issues during AI design and conception have been established. In response to this gap, we proposed the explainability in design (EID) methodological framework for addressing explainability problems in AI systems. We explored the literature on AI explainability to narrow down the field into six major explainability principles that will aid designers in brainstorming around the metrics and guide the critical thinking process. EID is a step-by-step guide to AI design that has been refined over a series of user studies and interviews with experts in AI explainability. It is devised for software design teams to uncover and resolve potential issues in their AI products and to simply refine and explore the explainability of their products and systems. The EID methodology is a novel framework that aids in the design and conception stages of the AI pipeline and can be integrated into the form of a step-by-step card game. Empirical studies involving AI system designers have shown that EID can decrease the barrier of entry and the time and experience required to effectively make well-informed decisions for integrating explainability into their AI solutions. © The author(s) 2023.

关键词： Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Fairness in Design: A Framework for Facilitating Ethical Artificial Intelligence Designs

引用

International Journal of Crowd science 2023年第1期7卷 32-39页

作者： Zhang, Jiehuang Shu, Ying Yu, Han School of Computer Science and Engineering Nanyang Technological University Singapore639798 Singapore Alibaba-NTU Singapore Joint Research Institute Singapore637335 Singapore

As Artificial Intelligence (AI) and Digital Transformation (DT) technologies become increasingly ubiquitous in modern society, the flaws in their designs are starting to attract attention. AI models have been shown to be susceptible to biases in the training data, especially against underrepresented groups. Although an increasing call for AI solution designers to take fairness into account, the field lacks a design methodology to help AI design teams of members from different backgrounds brainstorm and surface potential fairness issues during the design stage. To address this problem, we propose the Fairness in Design (FID) framework to help AI software designers surface and explore complex fairness-related issues, that otherwise can be overlooked. We explore literature in the field of fairness in AI to narrow down the field into ten major fairness principles, which assist designers in brainstorming around metrics and guide thinking processes about fairness. FID facilitates discussions among design team members, through a game-like approach that is based on a set of prompt cards, to identify and discuss potential concerns from the perspective of various stakeholders. Extensive user studies show that FID is effective at assisting participants in making better decisions about fairness, especially complex issues that involve algorithmic decisions. It has also been found to decrease the barrier of entry for software teams, in terms of the pre-requisite knowledge about fairness, to address fairness issues so that they can make more appropriate related design decisions. The FID methodological framework contributes a novel toolkit to aid in the design and conception process of AI systems, decrease barriers to entry, and assist critical thinking around complex issues surrounding algorithmic systems. The framework is integrated into a step-by-step card game for AI system designers to employ during the design and conception stage of the life-cycle process. FID is a unique dec

关键词： Software design

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：