检索结果-内蒙古大学图书馆

Proceedings of the 38th International Conference on Neural Information Processing systems

作者： Haolin Wang Xuefeng Liu Jianwei Niu Wenkai Guo Shaojie Tang State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing China State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing China and Zhongguancun Laboratory Beijing China Center for AI Business Innovation Department of Management Science and Systems University at Buffalo Buffalo New York

ISBN: (纸本)9798331314385

Federated learning is a distributed machine learning paradigm designed to protect user data privacy, which has been successfully implemented across various scenarios. In traditional federated learning, the entire parameter set of local models is updated and averaged in each training round. Although this full network update method maximizes knowledge acquisition and sharing for each model layer, it prevents the layers of the global model from cooperating effectively to complete the tasks of each client, a challenge we refer to as layer mismatch. This mismatch problem recurs after every parameter averaging, consequently slowing down model convergence and degrading overall performance. To address the layer mismatch issue, we introduce the FedPart method, which restricts model updates to either a single layer or a few layers during each communication round. Furthermore, to maintain the efficiency of knowledge acquisition and sharing, we develop several strategies to select trainable layers in each round, including sequential updating and multi-round cycle training. Our theoretical analysis and experimental results show that the FedPart method significantly outperforms traditional full-network update strategies, achieving faster convergence, greater accuracy, and reduced communication and computational overhead.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Lightweight Disaster Semantic Segmentation for UAV On-Device Intelligence

Lightweight Disaster Semantic Segmentation for UAV On-Device...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Hanbin Lee Gihwan Kim Sangho Ha Hyungshin Kim Department of Computer Science and Engineering Embedded Systems Laboratory (ESLAB) Chungnam National University Daejeon Republic of Korea

ISBN: (数字)9798350360325

ISBN: (纸本)9798350360332

Unmanned aerial vehicles (UAVs) are being used for monitoring natural disasters. To promptly assess the severity of the disaster, it is advantageous to analyze the disaster scenes with the on-board computer in real time. Various AI segmentation models with high accuracy are developed mainly for offline processing. They require significant memory capacity and computational power. However, on-device AI has the challenge of compressing high-precision models due to the limited memory size and the small computation power. In this research, we develop a lightweight disaster semantic segmentation model for UAV on-device intelligence. From a simple FANet as our baseline, we apply various optimization, such as compact backbone, lightweight attention block, quantization, knowledge distillation, and post processing. With our optimized model, we can reduce 84.2% of the inference time while achieving 0.5% increase in accuracy compared to the baseline model.

关键词： Accuracy Quantization (signal) Disasters Computational modeling Semantic segmentation Autonomous aerial vehicles Real-time systems

来源：评论

学校读者我要写书评

暂无评论

BinaryAI: Binary Software Composition Analysis via Intelligent Binary Source Code Matching 24

BinaryAI: Binary Software Composition Analysis via Intellige...

引用

44th ACM/IEEE International Conference on Software engineering, ICSE 2024

作者： Jiang, Ling An, Junwen Huang, Huihui Tang, Qiyi Nie, Sen Wu, Shi Zhang, Yuqun Southern University of Science and Technology Shenzhen China Tencent Security Keen Lab Shanghai China Research Institute of Trustworthy Autonomous Systems Southern University of Science and Technology Shenzhen China Southern University of Science and Technol-ogy Department of Computer Science and Engineering Shenzhen China Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation China

ISBN: (纸本)9798400702174

While third-party libraries (TPLs) are extensively reused to enhance productivity during software development, they can also introduce potential security risks such as vulnerability propagation. Software composition analysis (SCA), proposed to identify reused TPLs for reducing such risks, has become an essential procedure within modern DevSecOps. As one of the mainstream SCA techniques, binary-to-source SCA identifies the third-party source projects contained in binary files via binary source code matching, which is a major challenge in reverse engineering since binary and source code exhibit substantial disparities after compilation. The existing binary-to-source SCA techniques leverage basic syntactic features that suffer from redundancy and lack robustness in the large-scale TPL dataset, leading to inevitable false positives and compromised recall. To mitigate these limitations, we introduce BinaryAI, a novel binary-to-source SCA technique with two-phase binary source code matching to capture both syntactic and semantic code features. First, BinaryAI trains a transformer-based model to produce function-level embeddings and obtain similar source functions for each binary function accordingly. Then by applying the link-time locality to facilitate function matching, BinaryAI detects the reused TPLs based on the ratio of matched source functions. Our experimental results demonstrate the superior performance of BinaryAI in terms of binary source code matching and the downstream SCA task. Specifically, our embedding model outperforms the state-of-the-art model CodeCMR, i.e., achieving 22.54% recall@l and 0.34 MRR compared with 10.75% and 0.17 respectively. Additionally, BinaryAI outperforms all existing binary-to-source SCA tools in TPL detection, increasing the precision from 73.36% to 85.84% and recall from 59.81% to 64.98% compared with the well-recognized commercial SCA product Black Duck. E-https://*** © 2024 ACM.

关键词： Syntactics

来源：评论

学校读者我要写书评

暂无评论

Transformative effects of ChatGPT on modern education: Emerging Era of AI Chatbots

引用

Internet of Things and Cyber-Physical systems 2024年第1期4卷 19-23页

作者： Gill, Sukhpal Singh Xu, Minxian Patros, Panos Wu, Huaming Kaur, Rupinder Kaur, Kamalpreet Fuller, Stephanie Singh, Manmeet Arora, Priyansh Parlikad, Ajith Kumar Stankovski, Vlado Abraham, Ajith Ghosh, Soumya K. Lutfiyya, Hanan Kanhere, Salil S. Bahsoon, Rami Rana, Omer Dustdar, Schahram Sakellariou, Rizos Uhlig, Steve Buyya, Rajkumar School of Electronic Engineering and Computer Science Queen Mary University of London London United Kingdom Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China Raygun Performance Monitoring Wellington New Zealand Center for Applied Mathematics Tianjin University Tianjin China Department of Science Kings Education London United Kingdom Cymax Group Technologies BC Canada QM Academy Queen Mary University of London London United Kingdom Jackson School of Geosciences University of Texas at Austin AustinTX United States Centre for Climate Change Research Indian Institute of Tropical Meteorology Pune India Microsoft Hyderabad India Institute for Manufacturing Department of Engineering University of Cambridge Cambridge United Kingdom Faculty of Computer and Information Science University of Ljubljana Ljubljana Slovenia Machine Intelligence Research Labs AuburnWA United States Faculty of Computing and Data Science FLAME University Maharashtra Pune India Department of Computer Science and Engineering Indian Institute of Technology Kharagpur India Department of Computer Science University of Western Ontario London Canada Sydney Australia School of Computer Science University of Birmingham United Kingdom School of Computer Science and Informatics Cardiff University Cardiff United Kingdom Distributed Systems Group Vienna University of Technology Vienna Austria Department of Computer Science University of Manchester Oxford Road Manchester United Kingdom Laboratory School of Computing and Information Systems The University of Melbourne Australia

ChatGPT, an AI-based chatbot, offers coherent and useful replies based on analysis of large volumes of data. In this article, leading academics, scientists, distinguish researchers and engineers discuss the transformative effects of ChatGPT on modern education. This research discusses ChatGPT capabilities and its use in the education sector, identifies potential concerns and challenges. Our preliminary evaluation shows that ChatGPT perform differently in different subject areas including finance, coding, maths, and general public queries. While ChatGPT has the ability to help educators by creating instructional content, offering suggestions and acting as an online educator to learners by answering questions, transforming education through smartphones and IoT gadgets, and promoting group work, there are clear drawbacks in its use, such as the possibility of producing inaccurate or false data and circumventing duplicate content (plagiarism) detectors where originality is essential. The often reported "hallucinations" within GenerativeAI in general, and also relevant for ChatGPT, can render its use of limited benefit where accuracy is essential. What ChatGPT lacks is a stochastic measure to help provide sincere and sensitive communication with its users. Academic regulations and evaluation practices used in educational institutions need to be updated, should ChatGPT be used as a tool in education. To address the transformative effects of ChatGPT on the learning environment, educating teachers and students alike about its capabilities and limitations will be crucial. © 2023 The Authors

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

PCB Defect Detection Method Based on G-YOLOv8

PCB Defect Detection Method Based on G-YOLOv8

引用

2024 International Conference on Cyber-Physical Social Intelligence, ICCSI 2024

作者： Gao, Chao Bai, Mengting Lian, Xiaoqin Hu, Xiangyu Han, Shuangshuang Chen, Shichao Beijing Technology and Business University School of Computer and Artificial Intelligence Beijing China Chinese Academy of Sciences State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Beijing China University of Science and Technology Beijing School of Computer and Communication Engineering Beijing China Chinese Academy of Sciences Beijing Huairou Academy of Parallel Sensing State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Beijing China

ISBN: (纸本)9798350376739

In the electronics industry, PCB is crucial for ensuring the product quality. However, these defects of PCB generally have the characteristics of complex backgrounds, small sizes and similar features, resulting in the issues of high computational complexity or low detection precision of PCB defect detection algorithms based on deep learning. To resolve the aforementioned concerns, this paper proposed a lightweight PCB defect detection model named G-YOLOv8 which was based on the YOLOv8. The G-YOLOv8 can effectively reduce the computational complexity by integrating the Ghost Conv into the Backbone and Neck networks. Furthermore, the G-C2f module was introduced into Neck network to strengthen the capability of G-YOLOv8 to perceive the feature information at different scales and semantic levels, thus improving the detection performance of the model for the PCB defect. Meanwhile, the performance of G-YOLOv8 model was evaluated on the public data set. The experimental results show that the parameter number of G-YOLOv8 is 2.6 million, which is decreased by 89.7% and 13.3% compared with SSD and YOLOv8, respectively. The computational load of G-YOLOv8 is 7.5 GFLOPs, which is decreased by 93.5% and 7.4% compared with SSD and YOLOv8, respectively. The mAP@0.5 of G-YOLOv8 is 94.4%, which is increased by 4.8% and 1.3% compared with SSD and YOLOv8, respectively. Hence, in the task of PCB defect detection, the G-YOLOv8 model can reduce the computational complexity while preserving a high detection precision, which can meet the requirements of PCB defect detection in industrial scenarios. © 2024 IEEE.

关键词： Electronics industry

来源：评论

学校读者我要写书评

暂无评论

Bi-level Optimal Planning of Voltage Regulator in Distribution systems Considering Maximization of Incentive-based Photovoltaic Energy Integration

引用

CSEE Journal of Power and Energy systems 2023年第6期9卷 2008-2017页

作者： Xu Xu Youwei Jia Chun Sing Lai Minghao Wang Zhao Xu Department of Electrical and Electronic Engineering Southern University of Science and TechnologyShenzhenChina Department of Electrical Engineering The Hong Kong Polytechnic UniversityHung HomHong Kong SARChina Department of Electrical and Electronic Engineering University Key Laboratory of Advanced Wireless Communications of Guangdong ProvinceSouthern University of Science and TechnologyShenzhen 515100China Department of Electronic and Computer Engineering Brunel Institute of Power SystemsBrunel University LondonLondonUB83PH United Kingdom Department of Electrical Engineering Shenzhen Research InstituteThe Hong Kong Polytechnic UniversityHung HomHong Kong SARChina

This paper focuses on optimal voltage regulator(VR)planning to maximize the photovoltaic(PV)energy integration in distribution *** describe the amount of dynamic PV energy that can be integrated into the power system,the concept of PV accommodation capability(PVAC)is introduced and modeled with *** proposed planning model is formulated as a Benders decomposition based bi-level stochastic optimization *** the upper-level problem,VR planning decisions and PVAC are determined via mixed integer linear programming(MILP)before considering *** in the lower-level problem,the feasibility of first-level results is checked by critical network constraints(*** magnitude constraints and line capacity constraints)under uncertainties considered by time-varying loads and PV *** this paper,these uncertainties are represented in the form of operational scenarios,which are generated by the Gaussian copula theory and reduced by a well-studied backward-reduction *** modified IEEE 33-node distribution grid is utilized to verify the effectiveness of the proposed *** results demonstrate that a PV energy integration can be significantly enhanced after optimal voltage regulator planning.

关键词： Bi-level stochastic optimization problem critical network constraints photovoltaic energy integration uncertainties voltage regulator planning

来源：评论

学校读者我要写书评

暂无评论

Proposed Privacy Preservation Technique for Color Medical Images

引用

Intelligent Automation & Soft Computing 2023年第4期36卷 719-732页

作者： Walid El-Shafai Hayam A.Abd El-Hameed Noha A.El-Hag Ashraf A.M.Khalaf Naglaa F.Soliman Hussah Nasser AlEisa Fathi E.Abd El-Samie Department of Electronics and Electrical Communications Faculty of Electronic EngineeringMenoufia UniversityMenouf32952Egypt Security Engineering Laboratory Department of Computer SciencePrince Sultan UniversityRiyadh11586Saudi Arabia Electrical Communications Engineering Department Faculty of EngineeringMinia UniversityMinia61111Egypt Higher Institute of Commercial Science Management Information SystemsEl-Mahala El-KobraEgypt Department of Information Technology College of Computer and Information SciencesPrincess Nourah Bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Department of Computer Sciences College of Computer and Information SciencesPrincess Nourah Bint Abdulrahman UniversityRiyadhSaudi Arabia

Nowadays,the security of images or information is very *** paper introduces a proposed hybrid watermarking and encryption technique for increasing medical image ***,the secret medical image is encrypted using Advanced Encryption Standard(AES)***,the secret report of the patient is embedded into the encrypted secret medical image with the Least Signiﬁcant Bit(LSB)watermarking *** that,the encrypted secret medical image with the secret report is concealed in a cover medical image,using Kekre’s Median Codebook Generation(KMCG)***,the stego-image obtained is split into 16 ***,it is sent to the *** adopt this strategy to send the secret medical image and report over a network *** proposed technique is assessed with different encryption quality metrics including Peak Signal-to-Noise Ratio(PSNR),Correlation Coefﬁcient(Cr),Fea-ture Similarity Index Metric(FSIM),and Structural Similarity Index Metric(SSIM).Histogram estimation is used to conﬁrm the matching between the secret medical image before and after *** results demonstrate that the proposed technique achieves good performance with high quality of the received medical image and clear image details in a very short processing time.

关键词： LSB steganography AES algorithm KMCG algorithm

来源：评论

学校读者我要写书评

暂无评论

A Unified Analytical Method to Quantify Three Types of Fast Frequency Response from Inverter-based Resources 22

A Unified Analytical Method to Quantify Three Types of Fast ...

引用

22nd Wind and Solar Integration Workshop, WIW 2023

作者： Dong, Shuan Fang, Xin Tan, Jin Gao, Ningchao Cui, Xiaofan Hoke, Anderson Power Systems Energy Center National Renewable Energy Laboratory Golden United States Department of Electrical and Computer Engineering Mississippi State University Starkville United States Department of Electrical & Computer Engineering University of Denver Denver United States Department of Energy Science & Engineering Stanford University Stanford United States

ISBN: (纸本)9781839539824

With more inverter-based resources (IBRs), our power systems have lower frequency nadirs following N-1 contingencies, and undesired under-frequency load shedding (UFLS) can occur. To address this challenge, IBRs can be programmed to provide at least three types of fast frequency response (FFR), e.g., step response, proportional response (P/f droop response), and derivative response (synthetic inertia). However, these heterogeneous FFR challenge the study of power system frequency dynamics. Thus, this paper develops an analytical frequency nadir prediction method that allows for the consideration of all three potential forms of FFR provided by IBRs. The proposed method provides fast and accurate frequency nadir estimation after N-1 generation tripping contingencies. Our method is grounded on the closed-form solution for the frequency nadir, which is solved from the second-order system frequency response model considering the governor dynamics and three types of FFR. The simulation results in the IEEE 39-bus system with different types of FFR demonstrate that the proposed method provides an accurate and fast prediction of the frequency nadir under various disturbances. © Energynautics GmbH.

关键词： Frequency response

来源：评论

学校读者我要写书评

暂无评论

Sea-Land Segmentation of Remote Sensing Images Based on SDW-UNet

引用

computer systems science & engineering 2023年第5期45卷 1033-1045页

作者： Tianyu Liu Pengyu Liu Xiaowei Jia Shanji Chen Ying Ma Qian Gao The Information Department Beijing University of TechnologyBeijing 100124China School of Physics and Electronic Information Engineering Qinghai Minzu UniversityXining810000China Advanced Information Network Beijing Laboratory Beijing100124China Computational Intelligence and Intelligent Systems Beijing key Laboratory Beijing100124China Department of Computer Science University of Pittsburgh15260USA

Image segmentation of sea-land remote sensing images is of great importance for downstream applications including shoreline extraction,the monitoring of near-shore marine environment,and near-shore target *** mitigate large number of parameters and improve the segmentation accuracy,we propose a new Squeeze-Depth-Wise UNet(SDW-UNet)deep learning model for sea-land remote sensing image *** proposed SDW-UNet model leverages the squeeze-excitation and depth-wise separable convolution to construct new convolution modules,which enhance the model capacity in combining multiple channels and reduces the model *** further explore the effect of position-encoded information in NLP(Natural Language Processing)domain on sea-land segmentation *** have conducted extensive experiments to compare the proposed network with the mainstream segmentation network in terms of accuracy,the number of parameters and the time cost for *** test results on remote sensing data sets of Guam,Okinawa,Taiwan China,San Diego,and Diego Garcia demonstrate the effectiveness of SDW-UNet in recognizing different types of sea-land areas with a smaller number of parameters,reduces prediction time cost and improves performance over other mainstream segmentation *** also show that the position encoding can further improve the accuracy of model segmentation.

关键词： Sea-land segmentation UNet depth-wise separable convolution squeeze-excitation position encoding

来源：评论

学校读者我要写书评

暂无评论

Van der Waals integration inch-scale 2D MoSe_(2) layers on Si for highly-sensitive broadband photodetection and imaging

引用

Nano Research 2023年第8期16卷 11422-11429页

作者： Yupiao Wu Shuo-En Wu Jinjin Hei Longhui Zeng Pei Lin Zhifeng Shi Qingming Chen Xinjian Li Xuechao Yu Di Wu School of Physics and Microelectronics and Key Laboratory of Material PhysicsZhengzhou UniversityZhengzhou 450052China Department of Electrical and Computer Engineering University of California San DiegoLa JollaCA 92093USA School of Microelectronics Science and Technology Sun Yat-Sen UniversityZhuhai 519082China Key Laboratory of Multifunctional Nanomaterials and Smart Systems Suzhou Institute of Nano-Tech and Nano-BionicsChinese Academy of SciencesSuzhou 215123China

As one of the most promising materials for two-dimensional transition metal chalcogenides(2D TMDs),molybdenum diselenide(MoSe_(2))has great potential in photodetectors due to its excellent properties like tunable bandgap,high carrier mobility,and excellent air *** 2D MoSe_(2)-based photodetectors have been reported to exhibit admired performance,the large-area 2D MoSe_(2)layers are difficult to be achieved via conventional synthesis methods,which severely impedes its future ***,we present the controllable growth of large-area 2D MoSe_(2)layers over 3.5-inch with excellent homogeneity by a simple post-selenization ***,a high-quality n-MoSe_(2)/p-Si van der Waals(vdW)heterojunction device is in-situ fabricated by directly growing 2D n-MoSe_(2)layers on the patterned p-Si substrate,which shows a self-driven broadband photoresponse ranging from ultraviolet to mid-wave infrared with an impressive responsivity of 720.5 mA·W^(−1),a high specific detectivity of 10^(13) Jones,and a fast response time to follow nanosecond pulsed optical *** addition,thanks to the inch-level 2D MoSe_(2)layers,a 4×4 integrated heterojunction device array is achieved,which has demonstrated good uniformity and satisfying imaging *** large-area 2D MoSe_(2)layer and its heterojunction device array have great promise for high-performance photodetection and imaging applications in integrated optoelectronic systems.

关键词： molybdenum diselenide large-area synthesis broadband photodetector integrated device array van der Waals(vdW)heterojunction

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：