检索结果-内蒙古大学图书馆

Big Data Analytics and Practices (IBDAP), International Conference on

作者： Jimin Park Bongjun Kim Junho Jeong Dept. of computer science and engineering Dongguk University Seoul Rep. of Korea

ISBN: (数字)9798350391749

ISBN: (纸本)9798350391756

Skeleton-based human action recognition technology, based on a skeleton framework, is increasingly adopted in visual safety monitoring systems as it does not require exposure of personal identity information. Among various visual-based safety monitoring tasks, fall incidents can sometimes be fatal, emphasizing the need for accurately classifying human body activities and providing prompt assistance. While artificial intelligence has been applied to visual-based solutions for action recognition, accurately classifying actions remains challenging due to the lack of training data. Research has attempted to improve model performance using synthetic data, yet discussions on the relationship between the quality of skeleton data obtained from synthetic data and model performance have been limited. In this proposed study, we demonstrate how the quality of skeleton data used in fall detection model training affects the performance of fall detection. Therefore, it is expected that the results of this study will serve as valuable foundational material for improving the performance of skeleton-based fall detection models.

关键词： Analytical models Visualization Pose estimation Skeleton Data models Safety Fall detection

来源：评论

学校读者我要写书评

暂无评论

An Advanced MIMO Multicarrier Framework for NTN-Based 5G Broadcast/Multicast Transmissions

An Advanced MIMO Multicarrier Framework for NTN-Based 5G Bro...

引用

IEEE Metrology for Aerospace

作者： Talha Faizur Rahman Vuk Marojevic Claudio Sacchi Dept. of Electrical and Computer Engineering Mississippi State University Mississippi State MS Dept. of Information Engineering and Computer Science University of Trento Trento Italy Dept. of Electrical Engineering The University of New Mexico Albuquerque NM

ISBN: (数字)9798350385045

ISBN: (纸本)9798350385052

Satellites are teaming up with 5G, forming a non-terrestrial network (NTN), to support broadband applications over wide coverage areas. However, low latency and high reliability are the challenges in NTN which is addressed through the use of low-earth orbit (LEO) satellites. High transmission efficiency can be accomplished by employing multiple-input multiple-output (MIMO) antenna systems and multicarrier wave-forms. Conventional multicarrier techniques, such as Orthogonal Frequency Division Multiplexing (OFDM) and its variants, are sensitive to the nonlinear distortion of high power amplifiers (HPAs) and high Doppler environments. Therefore, we consider Constant Envelope OFDM (CE-OFDM) assisted with Space-Time Shift Keying (STSK) and Low-Density Parity Check (LDPC) encoding for 5G broadcast/multicast systems based on LEO satellites working in the L-band. We propose a low complexity receiver based on A Posteriori Probability (APP) STSK decoding integrated with LDPC decoding, in which we derive the APP of LDPC coded bits for the STSK-CE-OFDM transmission. We analyze the computational complexity of the proposed receiver scheme and conclude that it is effective in counteracting HPA nonlinearity and high Doppler effects inherent to LEO satellites. Numerical results show that the proposed framework outper-forms conventional multicarrier techniques with a gain as high as 16 dB in the presence of amplifier nonlinearity.

关键词： 5G mobile communication OFDM Satellite broadcasting Transmitting antennas Low earth orbit satellites Receivers Parity check codes

来源：评论

学校读者我要写书评

暂无评论

Sketch to Adapt: Fine-Tunable Sketches for Efficient LLM Adaptation

arXiv

引用

arXiv 2024年

作者： Zhang, Tianyi Su, Junda Desai, Aditya Wu, Oscar Xu, Zhaozhuo Shrivastava, Anshumali Dept. of Computer Science Rice University United States Dept. of Electrical Engineering and Computer Sciences University of California Berkeley United States xMAD.ai Dept. of Computer Science Stevens Institute of Technology United States

Adapting pre-trained large language models (LLMs) is crucial but challenging due to their enormous size. Parameter-efficient fine-tuning (PEFT) techniques typically employ additive adapters applied to frozen model weights. To further reduce memory usage, model weights can be compressed through quantization. However, existing PEFT methods often yield suboptimal model quality due to restrictive assumptions, such as imposing low-rank constraints on adapters to reduce trainable parameters. We find that sketching, a popular data compression technique, can serve as an efficient adaptation strategy for LLMs while avoiding low-rank assumptions. We introduce SketchTune, a compressive adaptation strategy that compresses LLM weights into compact fine-tunable sketches, integrating compression and adaptation into a unified framework. This integration eliminates the need for complex two-path computation common in existing PEFT techniques, enabling faster and more memory-efficient training and inference. SketchTune is supported by mathematical insights into matrix classes that are better approximated using sketching rather than low-rank methods. Our rigorous evaluations with Llama-1/2/3 models demonstrate that SketchTune outperforms leading PEFT methods across diverse tasks including math problem-solving, common sense reasoning, and instruction following, while using substantially smaller base models and comparable trainable parameters. As a highlight, SketchTune outperforms LoRA, DoRA, and S2FT on commonsense and math benchmarks using 2.6-3.5× smaller base models and exceeds LoftQ in accuracy by 14.48% on GSM8K with 7.3× fewer trainable parameters. © 2024, CC BY.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Explainable Adversarial Attacks on Coarse-to-Fine Classifiers

arXiv

引用

arXiv 2025年

作者： Heidarizadeh, Akram Hatfield, Connor Lazzarotto, Lorenzo Cai, HanQin Atia, George Dept. of Electrical and Computer Engineering University of Central Florida OrlandoFL United States School of Technology Pontifícia Universidade Católica do Rio Grande Sul Porto Alegre Brazil Dept. of Statistics and Data Science Dept. of Computer Science University of Central Florida OrlandoFL United States Dept. of Electrical and Computer Engineering Dept. of Computer Science University of Central Florida OrlandoFL United States

Traditional adversarial attacks typically aim to alter the predicted labels of input images by generating perturbations that are imperceptible to the human eye. However, these approaches often lack explainability. Moreover, most existing work on adversarial attacks focuses on single-stage classifiers, but multi-stage classifiers are largely unexplored. In this paper, we introduce instance-based adversarial attacks for multi-stage classifiers, leveraging Layer-wise Relevance Propagation (LRP), which assigns relevance scores to pixels based on their influence on classification outcomes. Our approach generates explainable adversarial perturbations by utilizing LRP to identify and target key features critical for both coarse and fine-grained classifications. Unlike conventional attacks, our method not only induces misclassification but also enhances the interpretability of the model’s behavior across classification stages, as demonstrated by experimental results. Copyright © 2025, The Authors. All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Synergistic Integration of Photovoltaic and the Grid: An Intelligent ANFIS-Governed Interleaved Boost Converter Approach

Synergistic Integration of Photovoltaic and the Grid: An Int...

引用

2024 International Conference on Advancement in Renewable Energy and Intelligent Systems, AREIS 2024

作者： Balaji, V. Ganga, M. Barathi, K. Gouda, Pramod Kumar Prasad, Rajesh Venkatesh, Mammula MAI-NEFHI College of Engineering and Technology Department of Electrical and Computer Engineering Asmara Eritrea Hindusthan College of Engineering and Technology Department of Biomedical Engineering Coimbatore India Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Dept. of Eee Tamil Nadu India Aditya Institute of Technology and Management Dept. of Eee Andhra Pradesh Srikakulam India Stella Mary's College of Engineering Dept. of Eee Tamil Nadu India Avanthi's Research and Technological Academy Dept. of Eee A. P Vizianagaram531162 India

ISBN: (纸本)9798350387230

The transition towards Renewable Energy Sources (RES), mainly solar power is crucial in addressing the depletion of fossil fuel-based energy supplies and mitigating environmental concerns. This paper proposes an innovative approach to enhance efficiency of PhotoVoltaic (PV) systems by integrating a two-phase Interleaved Boost (IB) converter with an Adaptive Neuro Fuzzy Inference System (ANFIS) controller. The ANFIS controller optimizes the operation of the IB converter by adjusting the duty cycle based on two key input parameters: solar irradiation and temperature. This adaptive control method effectively transforms solar energy into electrical power, maximizing the PV system's power production. By using the ANFIS mechanism, the suggested system improves total output power, decreases overshoot, and produces a higher voltage output than conventional techniques, all of which lead to increased performance and dependability. The IB converter operates with an efficiency of 96.15% and minimized settling time. © 2024 IEEE.

关键词： Adaptive control systems

来源：评论

学校读者我要写书评

暂无评论

An Improved YOLOv5 Detection Method for Strip Surface Defect 28

An Improved YOLOv5 Detection Method for Strip Surface Defect

引用

28th International Conference on Automation and Computing, ICAC 2023

作者： Liu, Yefeng Yu, Longfei Zhang, Qichun School of Automation and Electrical Dept. Liaoning Key Laboratory of Information Physical Fusion and Intelligent Cnc Machine Tool Manufacturing Shenyang Institute of Technology Dept. Fushun China Shenyang Ligong University Dept. School of Automation and Electrica Engineering China University of Bradford Bradford Dept. Department of Computer Science United Kingdom

ISBN: (纸本)9798350335859

Traditional steel surface defect detection methods have issues such as low detection accuracy and efficiency. It is suggested a technique based on enhanced YOLOv5 steel surface fault identification. In order to improve the network's capacity to handle feature information, this study suggests adding SE attention mechanisms to the YOLOv5s network's Backbone, Neck, and SPPF and replacing the C3 structure at the end of the Backbone with C3TR. To achieve network model lightweight, it is proposed to replace the Neck part in the YOLOv5s model and the convolution in SPPF with Ghost convolution. Then, to improve network feature fusion, extend the improved SPPF to the Neck part. Introduce the SPD convolution in the Backbone section to reduce the information loss caused by the neural network's convolution. The tested mAP is 75.3%, which is 5.3% higher than the original model, 24.6% higher than the SSD method, and 4.6% higher than the YOLOv6 algorithm after the upgraded model has been trained on the NEU-DET dataset. The observed FPS is 81.3, which is 6.3 more than the relative literature algorithm and 18.0 more than the SSD algorithm. © 2023 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Beyond the ROC curve: classification trees using cost-optimal curves, with application to imbalanced datasets 24

Beyond the ROC curve: classification trees using cost-optima...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Magzhan Gabidolla Arman Zharmagambetov Miguel Á. Carreira-Perpiñán Dept. of Computer Science and Engineering University of California Merced Meta AI (FAIR)

Important applications such as fraud or spam detection or churn prediction involve binary classification problems where the datasets are imbalanced and the cost of false positives greatly differs from the cost of false negatives. We focus on classification trees, in particular oblique trees, which subsume both the traditional axis-aligned trees and logistic regression, but are more accurate than both while providing interpretablemodels. Rather than using ROC curves, we advocate a loss based on minimizing the false negatives subject to a maximum false positive rate, which we prove to be equivalent to minimizing a weighted 0/1 loss. This yields a curve of classifiers that provably dominates the ROC curve, but is hard to optimize due to the 0/1 loss. We give the first algorithm that can iteratively update the tree parameters globally so that the weighted 0/1 loss decreases monotonically. Experiments on various datasets with class imbalance or class costs show this indeed dominates ROC-based classifiers and significantly improves over previous approaches to learn trees based on weighted purity criteria or over- or undersampling.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Career Craft AI: A Personalized Resume Analysis and Job Recommendations System

Career Craft AI: A Personalized Resume Analysis and Job Reco...

引用

Innovative Sustainable Technologies for Energy, Mechatronics, and Smart Systems (ISTEMS), International Conference on

作者： Akshay Rajput Akanksha Dubey Rishabh Thakur Deepak Singh Upendra Pratap Singh Dept. of Computer Science and Engineering Graphic Era Deeemed to be University Dehradun India

ISBN: (数字)9798350384246

ISBN: (纸本)9798350384253

In the today's era of endless career options, the challenge to get that one ideal job that aligns best with individual's skills and interests is absent due to lack of personalized recommendations and lack of guidance about resume making it becomes challenging to find best jobs and get shortlisted for the hiring process of a company. This paper introduces a novel approach to address these challenges by the development of a Career Crafter AI - A Resume analysis and Job Recommendation System. Knowing the limitations of existing recommendations systems that primarily rely on individuals provided interests, our system, implemented on the e-learning platform for skill building, Level Up, goes beyond the traditional approach of recommendation system. Using Vector Machine algorithm and the cosine similarity our system classifies the users based on their skills and profile mentioned on the platform and Natural Language Processing enables to provide the tips and recommendation to improve our resume and get the insights about the user for better recommendation. The recommendation engine utilizes these insights to provide accurate and best fit jobs to the users and as well as provide potential candidate for job roles to the recruiters This research paper aims to provide a one stop destination for learners to improve their resume for getting an edge during shortlisting and interviews and to transform and optimize the job recommendation systems by offering a more personalized and effective approach, empowering individuals to navigate their career paths with confidence and precision.

关键词： Support vector machines Electronic learning engineering profession Processor scheduling Resumes Transforms Vectors

来源：评论

学校读者我要写书评

暂无评论

Basic Study on Transmission Power Enhancement in Dynamic Wireless Power Transfer Systems using Circular–Solenoid Multi-coil

Basic Study on Transmission Power Enhancement in Dynamic Wir...

引用

Wireless Power Week (WPW)

作者： Harutaka Suzuki Ryosuke Ota Dept. of Electrical Engineering and Computer Science Tokyo Metropolitan University Tokyo Japan

ISBN: (数字)9798350349139

ISBN: (纸本)9798350349146

This paper proposes a method for stable power transmission in dynamic wireless power transfer (DWPT) systems with a coil array installed on the ground. The proposed method adopts a Circular-Solenoid type multi-coil on the electric vehicle (EV) side to utilize both vertical and horizontal magnetic fields, achieving stable power transmission all over the DWPT lane. In the experiments, a prototype was constructed, and the magnetic coupling characteristics of the ground and EV side coils were clarified. The power transmission characteristics were then derived, and a guideline was provided for efficiently and stably transmitting 3 kW of power to EVs all over the DWPT lane.

关键词： Coils Couplings Wireless communication Magnetic field measurement Power transmission Prototypes Wireless power transfer

来源：评论

学校读者我要写书评

暂无评论

Synergizing Insights for Precise Rice Leaf Disease Diagnosis Via Multi-Modal Fusion

Synergizing Insights for Precise Rice Leaf Disease Diagnosis...

引用

International Conference on Advanced Computing and Communication Systems (ICACCS)

作者： Shelly Chelsy Pally Jenefa J Dept. of Computer Science and Engineering CHRIST (Deemed to be University) engaluru Karnataka India

ISBN: (数字)9798350384369

ISBN: (纸本)9798350384376

Rice holds a significant position in India, especially in the southern part of the country, where people tend to eat some rice at least once a day. Farmers are facing a huge loss due to diseases in leaf, which is the main problem of agriculture. By using techniques like machine learning, main problems detection can be done. This review, discusses common plant diseases that affect the leaf. Some include Leaf Spots, Rusts, Fusarium Wilt, Early Blight, Powdery Mildew and Downey Mildew. Our research found that machine learning techniques on rice plants make finding diseases on leaves easier. Finally, we concluded that the most accurate method is the Enhanced VGG16, with an accuracy of 99.60% because it is really good at spotting diseases on rice leaves because it's great at recognizing the small details and patterns in leaf pictures. This helps it to tell the diseases apart more accurately and make fewer mistakes in identifying them.

关键词： Productivity Accuracy Reviews Prevention and mitigation Computational modeling Crops Prediction algorithms Medical diagnosis Reliability Diseases

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：