检索结果-内蒙古大学图书馆

Subspace Deep Learning of Independent Component Regression Model for Ensemble Spectroscopic Calibration

IEEE Sensors Journal 2025年

作者： Zheng, Junhua Zhou, Hansong Liu, Xinyu Lyu, Yuting Yang, Zeyu Ge, Zhiqiang Zhejiang University of Science and Technology School of Automation and Electrical Engineering Hangzhou China Huzhou University Huzhou Key Laboratory of Intelligent Sensing and Optimal Control for Industrial Systems School of Engineering Huzhou313000 China Southeast University School of Mathematics Nanjing210096 China

Independent component regression (ICR) has become as an important spectroscopic calibration modeling method, due to its advantages in extracting non-Gaussian and high-order statistic features. While the calibration performance of ICR was recently improved by its deep learning form, the result is unstable which is subjected to different random seeds in the model initialization stage. This, however, will make the model less trustworthy in practical industrial applications. In this paper, a subspace deep learning framework is formulated to combine the ideas of both deep learning and ensemble learning, in order to improve the prediction stability of the ICR model. Meanwhile, with an appropriate design of the ensemble learning strategy in the subspace modeling framework, the calibration performance can be further enhanced. For performance evaluation, a detailed benchmark case study on the Corn dataset is carried out. Compared to ordinary and deep residual ICR models, both of the calibration performance and prediction stability have been significantly improved by the subspace deep learning approach. © 2001-2012 IEEE.

关键词： Linear regression

来源：评论

学校读者我要写书评

暂无评论

Zero-shot Autonomous Microscopy for Scalable and Intelligent Characterization of 2D Materials

arXiv

引用

arXiv 2025年

作者： Yang, Jingyun Yin, Ruoyan Avery Jiang, Chi Hu, Yuepeng Zhu, Xiaokai Hu, Xingjian Kumar, Sutharsika Wang, Xiao Zhai, Xiaohua Rong, Keran Zhu, Yunyue Zhang, Tianyi Yin, Zongyou Kong, Jing Gong, Neil Zhenqiang Ren, Zhichu Wang, Haozhe Department of Electrical and Computer Engineering Duke University United States School of Computing National University of Singapore Singapore Google Deepmind United Kingdom Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology United States School of Chemistry Australian National University Australia Department of Materials Science and Engineering Massachusetts Institute of Technology United States

Characterization of atomic-scale materials traditionally requires human experts with months to years of specialized training. Even for trained human operators, accurate and reliable characterization remains challenging when examining newly discovered materials such as two-dimensional (2D) structures. This bottleneck drives demand for fully autonomous experimentation systems capable of comprehending research objectives without requiring large training datasets. In this work, we present ATOMIC (Autonomous technology for Optical Microscopy & Intelligent Characterization), an end-to-end framework that integrates foundation models to enable fully autonomous, zero-shot characterization of 2D materials. Our system integrates the vision foundation model (i.e., Segment Anything Model), large language models (i.e., ChatGPT), unsupervised clustering, and topological analysis to automate microscope control, sample scanning, image segmentation, and intelligent analysis through prompt engineering, eliminating the need for additional training. When analyzing typical MoS2 samples, our approach achieves 99.7% segmentation accuracy for single layer identification, which is equivalent to that of human experts. In addition, the integrated model is able to detect grain boundary slits that are challenging to identify with human eyes. Furthermore, the system retains robust accuracy despite variable conditions including defocus, color temperature fluctuations, and exposure variations. It is applicable to a broad spectrum of common 2D materials—including graphene, MoS2, WSe2, SnSe—regardless of whether they were fabricated via top-down or bottom-up methods. This work represents the implementation of foundation models to achieve autonomous analysis, establishing a scalable and data-efficient characterization paradigm that fundamentally transforms the approach to nanoscale materials research. © 2025, CC BY.

关键词： Layered semiconductors

来源：评论

学校读者我要写书评

暂无评论

Optimizing Cloud Security with CNN and XGBoost Models for Intrusion Detection Systems

Optimizing Cloud Security with CNN and XGBoost Models for In...

引用

Integrated Circuits and Communication Systems (ICICACS), IEEE International Conference on

作者： Amit Karbhari Mogal V. Anitha Preeti Nitin Bhatt Kriti Srivasatava S. Rukmani Devi Anvesh Perada Department of Computer Science and Application MVP Samaj's CMCS College Nashik India Department of Computer Science and Engineering School of Engineering and Technology Dhanalakshmi Srinivasan University Trichy India Department of Electronics & Communication Engineering IIMT College of Engineering Greater Noida India Department of Computer Science and Engineering (Data Science) Dwarkadas J Sanghvi College of Engineering Vile Parle India Department of Computer Science Saveetha College of Liberal Arts and Sciences SIMATS Deemed to be University Chennai India Department of Electrical and Computer Engineering Drexel University Philadelphia United States of America

ISBN: (数字)9798331508456

ISBN: (纸本)9798331508463

Cloud computing and Internet growth have simplified certain formerly difficult activities. This advancement has also revealed many security weaknesses. Due to cyberattacks, organizations need Intrusion Detection Systems (IDS) to protect their data and networks. Preprocessing, feature selection, and model training comprise the suggested strategy. Preprocessing includes data cleansing, standardization, and labeling. Information gain, chi-square, and PSO are used to choose features. CNN-XGBoost trains the model. The CNN-XGBoost model outperforms solo CNN and XGBoost with an average accuracy of 92.08%. This accuracy proves the hybrid method's cloud computing breach detection usefulness. The study underlines the importance of enhanced IDS models in cyber risk reduction. CNN's feature extraction and XGBoost's classification power provide a powerful, efficient, and reliable cloud intrusion detection model. Modern organizations benefit from this network security method.

关键词： Training Cloud computing Accuracy Computational modeling Standards organizations Intrusion detection Standardization Feature extraction Reliability Integrated circuit modeling

来源：评论

学校读者我要写书评

暂无评论

Uncertainty-Guided Cross Attention Ensemble Mean Teacher for Semi-Supervised Medical Image Segmentation

Uncertainty-Guided Cross Attention Ensemble Mean Teacher for...

引用

IEEE Workshop on Applications of computer Vision (WACV)

作者： Meghana Karri Amit Soni Arya Koushik Biswas Nicolò Gennaro Vedat Cicek Gorkem Durak Yuri S. Velichko Ulas Bagci Department of Radiology Northwestern University Chicago IL USA School of Computer Science Engineering and Technology Bennett University Greater Noida UP India

ISBN: (数字)9798331510831

ISBN: (纸本)9798331510848

This work proposes a novel framework, UncertaintyGuided Cross Attention Ensemble Mean Teacher (UGCEMT), for achieving state-of-the-art performance in semisupervised medical image segmentation. UG-CEMT leverages the strengths of co-training and knowledge distillation by combining a Cross-attention Ensemble Mean Teacher framework (CEMT) inspired by Vision Transformers (ViT) with uncertainty-guided consistency regularization and Sharpness-Aware Minimization emphasizing uncertainty. UG-CEMT improves semi-supervised performance while maintaining a consistent network architecture and task setting by fostering high disparity between sub-networks. Experiments demonstrate significant advantages over existing methods like Mean Teacher and Crosspseudo Supervision in terms of disparity, domain generalization, and medical image segmentation performance. UG-CEMT achieves state-of-the-art results on multi-center prostate MRI and cardiac MRI datasets, where object segmentation is particularly challenging. Our results show that using only 10% labeled data, UG-CEMT approaches the performance of fully supervised methods, demonstrating its effectiveness in exploiting unlabeled data for robust medical image segmentation. The code is publicly available at https://***/Meghnak13/UG-CEMT

关键词： Image segmentation computer vision Uncertainty Codes Magnetic resonance imaging Object segmentation Network architecture Transformers Minimization Biomedical imaging

来源：评论

学校读者我要写书评

暂无评论

Silicon-Based Single Photon Avalanche Diode Technologies: Structures, Performance, and Challenges

IEEE Electron Devices Reviews

引用

IEEE Electron Devices Reviews 2025年 2卷 69-97页

作者： Sumaiya Hoque Md Anas Abdullah Mrwan Alayed Mohamed Elamien M. Jamal Deen Department of Electrical and Computer Engineering McMaster University Hamilton ON Canada Microelectronics and Semiconductors Institute King Abdulaziz City for Science and Technology (KACST) Riyadh Saudi Arabia School of Biomedical Engineering McMaster University Hamilton ON Canada

Solid-state Single Photon Avalanche Diodes (SPADs) have emerged as versatile photodetectors, playing a pivotal role in cutting-edge technologies such as photon counting, timing, and imaging. Their exceptional characteristics — high timing resolution, fast response, high gain, excellent quantum efficiency, and immunity to magnetic fields — make them a compelling alternative to traditional Photomultiplier Tubes (PMTs). SPADs' unique ability to detect single photons, combined with their digital output and compact size, has positioned them as a superior choice for a wide range of applications in engineering, environmental monitoring, and healthcare. This review provides a comprehensive analysis of SPAD technology, focusing on its design principles, performance metrics, and applications. We begin by exploring the fundamental operating principles of SPADs and their key performance metrics. Next, we examine various SPAD structures across different technology nodes, offering a comparative analysis to highlight the latest advancements. Based on our review, we identify several techniques to enhance SPAD performance, including: doping profile engineering, such as using high-voltage layers and doping compensation layers, upgrading optical stacks, and integration of high-speed front-end circuitry. We then introduce application-specific figures-of-merit and assess their relevance to the reviewed structures. Additionally, we address emerging trends in SPAD design, such as leveraging mask layer technology to achieve significant advancements. Finally, we conclude with an overview of research challenges, potential solutions, and future directions in SPAD technology, providing valuable insights to drive further innovations in this dynamic field.

关键词： Single-photon avalanche diodes Photonics Silicon Absorption Timing Reviews Indium gallium arsenide CMOS technology Breakdown voltage P-n junctions

来源：评论

学校读者我要写书评

暂无评论

A multip-level additive distortion method for security improvement in palette image steganography

引用

Journal of Visual Communication and Image Representation 2025年 110卷

作者： Chen, Yi Wang, Hongxia Cui, Yunhe Shen, Guowei Guo, Chun Liu, Yong Wu, Hanzhou State Key Laboratory of Public Big Data College of Computer Science and Technology Guizhou University Guiyang550025 China Guizhou Provincial Characteristic Key Laboratory of Software Engineering and Information Security Guizhou University Guiyang550025 China School of Cyber Science and Engineering Sichuan University Chengdu610065 China School of Artificial Intelligence Hubei University of Automotive Technology Shiyan442002 China School of Big Data and Computer Science Guizhou Normal University Guiyang550025 China School of Communication and Information Engineering Shanghai University Shanghai200444 China

With the rapid development of the Internet and communication technology, palette images have become a preferred media for steganography. However, the security of palette image steganography faces a big problem. To address this, we propose a multiple-level additive distortion method for security improvement in palette image steganography. The proposed multiple-level additive distortion method comprises an index-level cost method and a pixel-level cost method. The index-level and the pixel-level costs by the two methods can respectively reflect the relationship changes of adjacent indices and the pixels corresponding to the adjacent indices. Meanwhile, the index-level and the pixel-level costs can also reflect the modification impact of steganography. Therefore, the proposed method can improve the security of palette image steganography. We conducted extensive experiments on three datasets to verify the security improvement. Experiment results have shown our proposed multiple-level distortion method indeed has an advantage in security when compared with four state-of-the-art methods. © 2025 Elsevier Inc.

关键词： Additive costs Color table Index matrix Multiple-level distortions Palette images Security Steganography

来源：评论

学校读者我要写书评

暂无评论

Fake News Detection and Classification Using Deep Learning Techniques for Social Media Text 6th

Fake News Detection and Classification Using Deep Learning T...

引用

6th International Conference on Advanced Informatics for Computing Research, ICAICR 2023

作者： Kumari Shalini, A. Saxena, Sameer Suresh Kumar, B. AMITY Institute of Information Technology Amity University Rajasthan Jaipur India School of Computer Science and Engineering Sanjay Ghodawat University Kolhapur India

ISBN: (纸本)9783031840616

The rise of social media networks has become a critical component of modern human living due to recent developments and advances in computer technology. The present context has evolved into a widely recognized medium for the distribution of information and news pertaining to a wide range of topics, including daily reports. This timeframe serves as the main period for data collecting and information transmission. This environment has a multitude of advantages;yet, it also entails a significant amount of misinformation and material that may potentially perplex both readers and users seeking accurate information. A significant concern associated with this system is to the limited availability of reliable information and real-world news pertaining to social media. An integrated system has been built that leverages several components of chain type of technology and natural language processing (NLP) to use machine learning techniques for the purpose of detecting false news and enhancing the accuracy of identifying fraudulent user accounts and postings. Various methodologies have been used to extract and choose features in order to provide effective training rules and appropriately assess the test classifier. A comprehensive empirical investigation demonstrates that the Twitter dataset is a useful resource for assessing the precision of fake news detection in classification tasks. The deep learning based RNN system demonstrates a high accuracy rate of 98.80% throughout both training and testing phases. This accuracy surpasses that attained by traditional machine learning methods such as SVM, ANN, and NB. Furthermore, the proposed methodology has the capability to detect superfluous information, a phenomenon that may be corroborated via several sources or by referencing current events. The prime purpose of this approach is to enhance the eminence of the material, identify any problematic sections within the text, and generate the original versions of those sections, all whi

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

Prophet: SSD Failure Analysis and Prediction Guided by Flash Reliability Characteristics in Data Centers

引用

IEEE Transactions on computers 2025年

作者： Song, Yunpeng Liang, Yujiong Liu, Jialin Shi, Liang East China Normal University MoE Engineering Research Center of Software/Hardware Co-Design Technology and Application Shanghai China East China Normal University School of Computer Science and Technology Shanghai China Huazhong University of Science and Technology Wuhan National Laboratory for Optoelectronics Wuhan China

Solid-state drives (SSDs) are massively deployed in various fields, especially in data centers, for their excellent cost-effectiveness. However, SSDs may fail due to their imperfect manufacturing processes, resulting in system-level failures and even downtime in data centers. This makes SSD failure prediction critical. Current studies focus on dealing with data missing, numerical normalization, and other statistical issues in using machine learning methods, but the consideration of the reliability characteristics of the underlying flash media of SSDs and the timeliness (time duration between predicted failure and real failure) of SSD failure prediction result is missing. In this work, we study the failure characteristics of over 200,000 drives from industry data centers over a 4-year period, as well as daily data. The relationship between SSD attribute values and failures is first investigated. Then, we analyzed the SSD failure characteristics from several aspects (causes, differences between failures, and timeliness of prediction results) relying on flash reliability characteristics. Based on these, a novel SSD failure prediction method (Prophet) is proposed. Specifically, Prophet contains the following two components. First, to cope with the differences between failures, a diff-state method is proposed for differential machine learning modeling of SSDs in different "States". We define the "State" of an SSD, which represents the range of values in which the SSD currently lies in terms of some key attributes. Through flash reliability characteristics, we distinguish between different failures before training the model to obtain accurate predictions of different failure behaviors. Second, a recovery period method is proposed to enhance the timeliness of SSD failure prediction result by designing the sample selection method. The enhanced timeliness can be utilized by operations personnel to handle failed SSDs, such as replacement and repair. The evaluation results of

关键词： Failure rate

来源：评论

学校读者我要写书评

暂无评论

Fed-PID: An Adaptive Learning Rate Scheduler for Federated Learning with PID Controllers

引用

IEEE Transactions on Cognitive Communications and Networking 2025年

作者： Li, Shaofan Dai, Mingjun Kianoush, Sanaz Minora, Alberto Savazzi, Stefano Shenzhen University College of Electronics and Information Engineering China Kashi University School of Computer Science and Technology India Italy Bicocca University Milan Italy Cognimade srl Melzo Italy

The adaptation of learning rate can significantly enhance the performance of deep neural networks, especially in intricate environment of federated learning (FL) with diverse client settings. Therefore, the hyperparameter optimization is a critical component in FL. Distributed iterative optimization algorithms for FL encounter issues such as slow convergence due to inter-node coupling, as well as diminished inference accuracy. Furthermore, the reliance on laborious empirical learning rate adjustment results in weak adaptability to changing environments and slow responsiveness to dynamic conditions. In this work, we propose Fed-PID, a novel PID controller based adaptive learning rate scheduler for FL. Fed-PID employs distributed feedback control and PID technique to re-express distributed iterative optimization algorithms as distributed PID controllers. By equivalently transforming multiple coupled control systems into an extended dimensional error system, it converts the federated aggregation convergence problem into a consensus control problem. We design Server Side Parameter Scheduler (SSPS) to automatically adjusts the P, I, and D parameters for all clients by minimizing the extended dimensional error system from a global perspective and design Client Side Parameter Schedulers (CSPS) to automatically tune hyperparameters during local training from a micro perspective. Extensive experimental results on three public datasets and our self-built cobot radar localization dataset demonstrate that Fed-PID consistently outperforms FedAvg and other baseline methods. In particular, Fed-PID outperforms its competitors by 3.6% accuracy on CIFAR-10 and 5.85% on Cobot dataset under the Non-IID setting. © 2015 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Rheological performance regulation and material optimization of manufactured sand concrete in 3D printing

引用

Construction and Building Materials 2025年 483卷

作者： Zhu, Wenxuan Liu, Chao Zhang, Yu Zhang, Yunsheng Wang, Yifei Wang, Tilin School of Civil Engineering Xi'an University of Architecture and Technology Xi'an710055 China School of Civil Engineering Lanzhou University of Technology Gansu Advanced Civil Engineering Materials Engineering Research Center Lanzhou730050 China School of Materials Science and Engineering Southeast University Nanjing211189 China

This study systematically explores the rheological properties and printability of 3D printed manufactured sand concrete (3DPMSC), aiming to optimize the use of manufactured sand in 3D printed concrete (3DPC). The effects of water-binder ratio (W/B), superplasticizer (PCE) dosage, and cellulose ether (HPMC) content on the yield stress, plastic viscosity, and thixotropy of concrete were comprehensively analyzed through rheological and printing experiments. A secondary mixing process based on the cement-coated sand method, combined with a gradient PCE incorporation technique, was proposed to effectively mitigate the time-dependent rheological degradation of manufactured sand concrete. The results showed that a low W/B significantly enhanced the structural stability and thixotropy of the concrete while reducing its flowability. HPMC enhanced buildability by strengthening the particle network structure, and an appropriate dosage of PCE achieved a dynamic balance between shear thinning and thixotropic recovery, optimizing the printing quality. A material optimization design framework specifically for 3DPMSC is ultimately established, offering theoretical and practical guidance for sustainable use of manufactured sand in construction. © 2025 Elsevier Ltd

关键词： Shear thinning

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：