检索结果-内蒙古大学图书馆

Enhanced-Performance Self-Driven Infrared Heterojunction Photodetectors by Inserting a Thin Layer of Silver Nanoparticles Close to the Heterojunction Interface

SSRN

引用

SSRN 2023年

作者： Sun, Feiyang Yang, Shengyi Zhang, Zhenheng Sulaman, Muhammad Ge, Zhenhua Hu, Jinming Li, Chunyang Wu, Ying Liu, Xiaoxuan Zou, Bingsuo Beijing Key Lab of Nanophotonics and Ultrafine Optoelectronic Systems School of Physics Beijing Institute of Technology Beijing100081 China Centre for Artificial-Intelligence Nanophotonics School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology Shanghai200093 China School of Mechatronical Engineering Beijing Institute of Technology Beijing100081 China School of Physical Science and Technology Guangxi University Nanning530004 China

Lead selenide colloidal quantum dots (CQDs) are widely used as in infrared photodetectors due to their size-dependent bandgap tunability, facile solution-process and low cost. However, the low carrier mobility seriously affects the device performance. In order to increase the separation ratio and transport efficiency of carriers, heterojunctions are usually introduced to construct device. In this paper, a solution-processed self-driven heterojunction photodetector ITO/ZnO/PbSe/Ag is fabricated, in which the ZnO layer works as the electron-transporting and hole-blocking layer. By inserting a thin layer of Ag nanoparticles (NPs) into ZnO ETL but close to the ZnO/PbSe interface, the device performance was enhanced greatly. As a result, the responsivity of the heterojunction photodetector ITO/ZnO(80nm):Ag(10nm):ZnO(20nm)/ PbSe(300nm)/Ag reaches to 1.6 mA/W with a specific detectivity D* of 1.33 Jones under 8.5 1550 nm illumination under zero bias, showing an responsivity enhancement of 17-folds as compared with that of the control photodetector without inserting the thin layer of Ag NPs. Further, the underlain physical mechanisms for the enhanced-performance were discussed in details. In this way, it provides an efficient method for enhanced-performance self-driven infrared photodetectors by inserting a thin layer of metal NPs close to the heterojunction interface. © 2023, The Authors. All rights reserved.

关键词： Silver nanoparticles

来源：评论

学校读者我要写书评

暂无评论

A Survey on Fairness in Large Language Models

arXiv

引用

arXiv 2023年

作者： Li, Yingji Du, Mengnan Song, Rui Wang, Xin Wang, Ying College of Computer Science and Technology Jilin University Changchun130012 China Department of Data Science New Jersey Institute of Technology Newark United States School of Artificial Intelligence Jilin University Changchun130012 China Key Laboratory of Symbol Computation and Knowledge Engineering Jilin University Ministry of Education Changchun130012 China

Large Language Models (LLMs) have shown powerful performance and development prospects and are widely deployed in the real world. However, LLMs can capture social biases from unprocessed training data and propagate the biases to downstream tasks. Unfair LLM systems have undesirable social impacts and potential harms. In this paper, we provide a comprehensive review of related research on fairness in LLMs. Considering the influence of parameter magnitude and training paradigm on research strategy, we divide existing fairness research into oriented to medium-sized LLMs under pre-training and fine-tuning paradigms and oriented to large-sized LLMs under prompting paradigms. First, for medium-sized LLMs, we introduce evaluation metrics and debiasing methods from the perspectives of intrinsic bias and extrinsic bias, respectively. Then, for large-sized LLMs, we introduce recent fairness research, including fairness evaluation, reasons for bias, and debiasing methods. Finally, we discuss and provide insight on the challenges and future directions for the development of fairness in LLMs. Copyright © 2023, The Authors. All rights reserved.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

A Literature Review of Machine Learning Techniques for Recommender System In E-Commerce 5

A Literature Review of Machine Learning Techniques for Recom...

引用

5th IEEE International Conference on Electro-Computing Technologies for Humanity, NIGERCON 2024

作者： Ajibade, Samuel-Soma M. Jasser, Muhammed Basheer Hoong, Angela Lee Siew Culpable, Rex V. Quirante, Celestino A. Dacillo, Arvie L. Adetunla, Adedotun O. Akintoye, Kayode Akinlekan Sunway University School of Engineering and Technology Department of Data Science and Artificial Intelligence No. 5 Jalan Universiti Selangor Darul Ehsan Bandar Sunway47500 Malaysia Selangor Malaysia Cebu Technological University College of Technology Cebu Moalboal Philippines Afe Babalola University Dept. of Mechanical and Mechatronics Engineering Ekiti Ado Nigeria University of Johannesburg Dept. of Mechanical Engineering Science Johannesburg South Africa The Federal Polytechnic Department of Computer Science Ekiti State Ado Ekiti Nigeria

ISBN: (纸本)9798331542559

The swift expansion of e-commerce has prompted the creation of advanced recommender systems to augment user experience and boost sales. Machine learning techniques have become fundamental for the development of these systems, allowing firms to provide tailored product recommendations based on customer preferences and habits. This research presents a thorough literature assessment of machine learning methodologies utilized in recommender systems, emphasizing their predictive efficacy in the e-commerce sector. We analyze essential methodologies, including collaborative filtering, content-based filtering, hybrid models, and sophisticated techniques such as deep learning. Through the examination of 38 research in the domain, we elucidate the limitations, advantages, and possibilities of machine learning algorithms in delivering precise and scalable suggestions. Our research indicates that hybrid models, which integrate many recommendation strategies, demonstrate significant potential in addressing challenges like the cold start problem and data sparsity. The report delineates prospective research avenues, underscoring the necessity for continued innovation in the integration of machine learning into recommender systems to improve accuracy and customer happiness in e-commerce settings. © 2024 IEEE.

关键词： Marketplaces

来源：评论

学校读者我要写书评

暂无评论

OFFLINE MODEL-BASED OPTIMIZATION BY LEARNING TO RANK

arXiv

引用

arXiv 2024年

作者： Tan, Rong-Xi Xue, Ke Lyu, Shen-Huan Shang, Haopu Wang, Yao Wang, Yaoyuan Fu, Sheng Qian, Chao National Key Laboratory for Novel Software Technology Nanjing University China School of Artificial Intelligence Nanjing University China Key Laboratory of Water Big Data Technology of Ministry of Water Resources Hohai University China College of Computer Science and Software Engineering Hohai University China Huawei Technologies Ltd. China

Offline model-based optimization (MBO) aims to identify a design that maximizes a black-box function using only a fixed, pre-collected dataset of designs and their corresponding scores. This problem has garnered significant attention from both scientific and industrial domains. A common approach in offline MBO is to train a regression-based surrogate model by minimizing mean squared error (MSE) and then find the best design within this surrogate model by different optimizers (e.g., gradient ascent). However, a critical challenge is the risk of out-of-distribution errors, i.e., the surrogate model may typically overestimate the scores and mislead the optimizers into suboptimal regions. Prior works have attempted to address this issue in various ways, such as using regularization techniques and ensemble learning to enhance the robustness of the model, but it still remains. In this paper, we argue that regression models trained with MSE are not well-aligned with the primary goal of offline MBO, which is to select promising designs rather than to predict their scores precisely. Notably, if a surrogate model can maintain the order of candidate designs based on their relative score relationships, it can produce the best designs even without precise predictions. To validate it, we conduct experiments to compare the relationship between the quality of the final designs and MSE, finding that the correlation is really very weak. In contrast, a metric that measures order-maintaining quality shows a significantly stronger correlation. Based on this observation, we propose learning a ranking-based model that leverages learning to rank techniques to prioritize promising designs based on their relative scores. We show that the generalization error on ranking loss can be well bounded. Empirical results across diverse tasks demonstrate the superior performance of our proposed ranking-based models than twenty existing methods. Copyright © 2024, The Authors. All rights reserved.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

A Molecular Simulation Study on Pore-Scale Behaviour of Nitrogen-Based Fracking Fluids for Potential Geo-Energy Applications

SSRN

引用

SSRN 2024年

作者： Singh, Aparna Sengupta, Angan Roy, Debanjan Guha Department of Civil and Infrastructure Engineering Indian Institute of Technology Rajasthan Jodhpur342030 India Department of Chemical Engineering Affiliated faculty in School of Artificial Intelligence and Data Science Indian Institute of Technology Rajasthan Jodhpur342030 India Department of Civil and Infrastructure Engineering Affiliated faculty in Rishabh Centre for Research and Innovation in Clean Energy Indian Institute of Technology Rajasthan Jodhpur342030 India

Molecular simulations are efficient tools in differentiating individual effects of fluid-fluid interactions and pore-fluid interactions on thermophysical properties of confined fluids;e.g. the molecular packing, adsorption mechanics and availability of accessible pore volume for confined fluids and therefore, indicate the rock fracturing phenomena as a function of geological conditions, fracking fluids nature, its composition and rock mineralogy. Presently, we have deployed the GCMC molecular simulations to quantify the adsorption of pure nitrogen and N2-H2O mixture (50%-50% and 30%-70%) inside porous silica rocks. While we found that adsorption and molecular packing of pure nitrogen inside silica slit pores are only a function of pore height, which quantifies the pore-fluid interactions;however, for N2-H2O mixture adsorption and molecular packing of N2 inside silica slit pores has been additionally affected by the water content in the equilibrium bulk mixture that as well describes fluid-fluid interactions inside pores. It is interestingly noted that water in N2-H2O mixture results in water-assisted nitrogen adsorption inside hydrophilic silica slit pores, which has been further proven through the radial distribution function data calculations inside each slit pore. Also, the hydrophilic nature of silica increases water adsorption and hence reduces N2 adsorption inside the smallest pore of H = 20 Å. Such a reduction in N2 adsorption density below its bulk density without layering effect, inside 20 Å pore, further initiates the possibility of negative excess adsorption density. © 2024, The Authors. All rights reserved.

关键词： Nitrogen

来源：评论

学校读者我要写书评

暂无评论

Riemannian Self-Attention Mechanism for SPD Networks

arXiv

引用

arXiv 2023年

作者： Wang, Rui Wu, Xiao-Jun Li, Hui Kittler, Josef School of Artificial Intelligence and Computer Science Jiangnan University Wuxi214122 China Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence Jiangnan University China Centre for Vision Speech and Signal Processing University of Surrey GuildfordGU2 7XH United Kingdom

Symmetric positive definite (SPD) matrix has been demonstrated to be an effective feature descriptor in many scientific areas, as it can encode spatiotemporal statistics of the data adequately on a curved Riemannian manifold, i.e., SPD manifold. Although there are many different ways to design network architectures for SPD matrix nonlinear learning, very few solutions explicitly mine the geometrical dependencies of features at different layers. Motivated by the great success of self-attention mechanism in capturing long-range relationships, an SPD manifold self-attention mechanism (SMSA) is proposed in this paper using some manifold-valued geometric operations, mainly the Riemannian metric, Riemannian mean, and Riemannian optimization. Then, an SMSA-based geometric learning module (SMSA-GLM) is designed for the sake of improving the discrimination of the generated deep structured representations. Extensive experimental results achieved on three benchmarking datasets show that our modification against the baseline network further alleviates the information degradation problem and leads to improved accuracy. Copyright © 2023, The Authors. All rights reserved.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Extended Upscale and Downscale Representation with Cascade Arrangement

Extended Upscale and Downscale Representation with Cascade A...

引用

IEEE/SP Workshop on Statistical Signal Processing (SSP)

作者： Quang Manh Doan Tran Hiep Dinh Nguyen Linh Trung Diep N. Nguyen Avinash Kumar Singh Chin-Teng Lin Advanced Institute of Engineering and Technology VNU University of Engineering and Technology Hanoi Vietnam VNU University of Engineering and Technology Hanoi Vietnam UTS-VNU Joint Technology Research Centre Hanoi Vietnam School of Electrical and Data Engineeering Faculty of Engineering and Information Technology University of Technology Sydney Sydney Australia Australian Artificial Intelligence Institute School of Computer Science Faculty of Engineering and Information Technology University of Technology Sydney Sydney Australia

Smoothing filters are widely used in EEG signal processing for noise removal while preserving important features. Unlike common approaches in the time domain, a recent effective algorithm using the Upscale and Downscale Representation (UDR) technique has been introduced to process the signal in the image domain. The idea of UDR is to visualize the input with an appropriate line width, convert it to a binary image, and then smooth it by skeletonizing the signal object to a unit width and projecting it back to the time domain. We propose in this paper a cascaded UDR (CUDR) where the interested signal is filtered twice. CUDR’s performance is verified on simulated data with added white Gaussian noise and compared with the cascaded arrangement of some conventional techniques. Experimental results have demonstrated the outperformance of CUDR in terms of the fitting error when dealing with noisy signals, especially at a low signal-to-noise ratio.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Sequence number based secure routing algorithm for IoT networks

引用

Materials Today: Proceedings 2021年

作者： Kothandaraman, D. Naik Korra, Seena Balasundaram, A. Magesh Kumar, S. School of Computer Science and Artificial Intelligence S R University TS Warangal India School of Computer Science and Engineering Centre for Cyber Physical Systems VIT TN Chennai India Department of Computer Science and Engineering SIMATS TN Chennai India

Internet of Things (IoT) is one among the rapid growing networks, commonly used in smart mobile devices such as laptops, tablets, smartphones, etc. Due to the dynamic nature and partial resource availability of the IoT nodes, its communication links between intermediate relay nodes may fail frequently and this will affect the routing performance of the network. While designing secure routing algorithm for IoT networks, limits in existing proactive and reactive routing protocols creates challenges. Also due to Mobility, Bandwidth, Energy Constraints, Routing cost and Delay, the availability of the nodes may vary from node to node and thus causing an issue in identifying the positions of mobile malicious nodes. To overcome these issues, in the present paper Sequence Number based Secure Routing (SNSR) algorithm has been implemented using existing IoT packet structure. The objective of this paper is to maximizes the packet delivery ratio and network life-time in order to improve the network performance. On comparing simulation results of SNSR algorithm with AODV and ABR routing algorithm, SNSR algorithm produced better performance of 95% than others. © 2020

关键词： Network performance

来源：评论

学校读者我要写书评

暂无评论

Correction: Agile meets quantum: a novel genetic algorithm model for predicting the success of quantum software development project

引用

Automated Software engineering 2024年第1期32卷 1-11页

作者： Khan, Arif Ali Akbar, Muhammad Azeem Lahtinen, Valtteri Paavola, Marko Niazi, Mahmood Alatawi, Mohammed Naif Alotaibi, Shoayee Dlaim M3S Empirical Software Engineering Research Unit University of Oulu Oulu Finland Software Engineering Department Lappeenranta-Lahti University of Technology Lappeenranta Finland Quanscient Oy Tampere Finland VTT Technical Research Ctr of Finland Ltd Oulu Finland Department of Information and Computer Science King Fahd University of Petroleum and Minerals Dhahran Saudi Arabia Interdisciplinary Research Centre for Intelligent Secure Systems King Fahd University of Petroleum and Minerals Dhahran Saudi Arabia Information Technology Department Faculty of Computers and Information Technology University of Tabuk Tabuk Saudi Arabia Department of Artificial Intelligence and Data Science College of Computer Science and Engineering University of Hail Hail Saudi Arabia

来源：评论

学校读者我要写书评

暂无评论

Analyze the anomalous behavior of wireless networking using the big data analytics

引用

Measurement: Sensors 2022年 23卷

作者： Abd Algani, Yousef Methkal Vinodhini, G Arul Freeda Isabels, K. Ruth Kaur, Chamandeep Treve, Mark Kiran Bala, B. Balaji, S. Devi, G. Usha Department of Mathematics Sakhnin College Israel Department of Mathematics The Arab Academic College for Education in Israel-Haifa Israel Department of Science and Humanities Saveetha School of Engineering SIMATS Chennai60210 India Department of Mathematics Saveetha Engineering College Saveetha Nagar Thandalam Tamilnadu Chennai602 105 India Computer Science Department Jazan University Jizan Saudi Arabia School of Languages and General Education Walailak University Nakhon Si Thammarat Thailand Head of the Department Department of Artificial Intelligence and Data Science K.Ramakrishnan College of Engineering Samayapuram Trichy621 112 India Department of Computer Science and Engineering Panimalar Engineering College India Department of Artificial Intelligence and Data Science K.Ramakrishnan College of Engineering Samayapuram Trichy India

Internet connections and cellular technologies are extensively used throughout the globe. Anomaly detection systems have considered an essential tool for detecting a broad range of hostile activity in the cyberspace domain. The researchers of this paper address the problems and existing knowledge of anomalous detecting for mobile networks as they prepare to embrace the "big data" age. As new computer cyber-security defects and vulnerabilities are reported every day, anomaly detection systems (ADSs) are getting increasingly crucial. The major objective is to develop methods for scanning networks activity and detecting unusual behaviours that could be the result of anomalous *** Dirichlet mixture prototype dependent on anomaly detection methodology is a proposed methodology called DM-ADs;anomaly detecting engine that incorporates 3 components: collecting and logging, pre-processing, and a novel statistical decision processor. This paper offers a hybrid anomaly detection method that combines several characteristic selecting strategies with an appropriate mixture approach to recognize each assault form with great precision. The suggested method's effectiveness is assessed using two databases, the NSL-KDD. The effectiveness of the suggested ADS was proved by retaining excellent precision and minimal false-positive percentages in all sorts of attacks. © 2022 The Authors

关键词： data Analytics

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：