检索结果-内蒙古大学图书馆

27th International Conference on Pattern Recognition, ICPR 2024

作者： Jiang, Hao Fang, Zhijun Shao, Xuan Jiang, Xiaoyan Hwang, Jenq-Neng School of Computer Science and Technology Donghua University Shanghai China School of Electronic and Electrical Engineering Shanghai University of Engineering Science Shanghai China Department of Electrical and Computer Engineering University of Washington SeattleWA United States

ISBN: (纸本)9783031784439

Lightweight yet reliable depth estimation models that can deployed on edge devices are crucial for the practical application of fields such as autonomous driving, robot navigation, and augmented reality. However, previous research often fails to balance accuracy and complexity well. Existing lightweight models still have relatively high error rates in specific scenarios, which makes them unsuitable for industrial applications. Hence, we propose LiDUT-Depth, a lightweight architecture for self-supervised monocular depth estimation that is enhanced through dynamic upsampling and triplet loss optimization, which achieves competitive results with smaller model sizes and lower computational complexity. Specifically, an efficient dynamic upsampling module (EDU Module) is employed to preserve detailed local features, thereby obtaining more accurate depth estimation results. In addition, an improved semantic-aware triplet loss (SaTri Loss) is integrated into the training loss, significantly improving depth estimation accuracy in gradient-rich regions. Experiments show that our architecture achieves a considerably enhanced accuracy compared with previous works with a much lower model size. Our codes and models are available here. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Augmented reality

来源：评论

学校读者我要写书评

暂无评论

Empirical Study on Data Attributes Insufficiency of Evaluation Benchmarks for LLMs 31

Empirical Study on Data Attributes Insufficiency of Evaluati...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Liu, Chuang Jin, Renren Yao, Zheng Li, Tianyi Cheng, Liang Steedman, Mark Xiong, Deyi College of Intelligence and Computing Tianjin University China School of Electrical Engineering and Computer Science University of Queensland Australia School of Informatics University of Edinburgh United Kingdom

ISBN: (纸本)9798891761964

Previous benchmarks for evaluating large language models (LLMs) have primarily emphasized quantitative metrics, such as data volume. However, this focus may neglect key qualitative data attributes that can significantly impact the final rankings of LLMs, resulting in unreliable leaderboards. In this paper, we investigate whether current LLM benchmarks adequately consider these data attributes. We specifically examine three attributes: diversity, redundancy, and difficulty. To explore these attributes, we propose a framework with three separate modules, each designed to assess one of the attributes. Using a method that progressively incorporates these attributes, we analyze their influence on the benchmark. Our experimental results reveal a meaningful correlation between LLM rankings on the revised benchmark and the original benchmark when these attributes are accounted for. These findings indicate that existing benchmarks often fail to meet all three criteria, highlighting a lack of consideration for multifaceted data attributes in current evaluation datasets. © 2025 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Machine learning model for predicting long-term energy consumption in buildings

引用

Discover Internet of Things 2025年第1期5卷 1-24页

作者： Hussien, Aseel Maksoud, Aref Al-Dahhan, Aisha Abdeen, Ahmed Baker, Thar Department of Architectural Engineering University of Sharjah Sharjah United Arab Emirates Department of Computer Science University of Sharjah Sharjah United Arab Emirates Civil Mining Environmental and Architectural Engineering University of Wollongong Wollongong Australia School of Architecture Technology and Engineering University of Brighton Brighton United Kingdom

The rapid growth in the construction sector has led to increased energy consumption and carbon emissions. Calculating energy usage and emissions is essential to energy security and promoting sustainable sector development. Therefore, the study objective is to investigate the utilazation of machine learning algorithm to predict long-term energy consumption in buildings sector, aiming to improve sustainable design and energy optimization, via the implementation of three machine learning models, XGBoost, Support Vector Regression, and Long-Short-Term Memory networks, to predict energy consumption. These models are adept at capturing complex interactions between building characteristics, environmental factors, and energy patterns. Although previous studies have explored various machine learning techniques for energy efficiency, limited research links these models to practical applications in building performance simulation. Furthermore, there is a lack of comparative evaluation of advanced machine learning models such as XGBoost, Support Vector Regression, and Long-Short-Term Memory to predict the energy consumption of building envelopes, particularly in hot climates such as the UAE. This research aims to fill this gap by providing a detailed comparison of these models against alternative approaches mentioned in the literature. The findings position Long-Short-Term Memory as a transformative force in predictive modeling, demonstrating exceptional precision with an R-squared value of 0.993 and a Mean Squared Error of 0.004. In contrast, Support Vector Regression and XGBoost showed limited predictive capabilities, with R-squared values of 0.462 and 0.94, respectively. This study establishes a solid data-driven foundation for architects and engineers to inform decisions on energy-efficient building designs, advocating Long-Short-Term Memory as the superior model for predicting energy performance. © The Author(s) 2025.

关键词： Green buildings

来源：评论

学校读者我要写书评

暂无评论

Numerical Study of Liquid Ammonia Flow Cooling of Power Electronics for Electric Aviation

Numerical Study of Liquid Ammonia Flow Cooling of Power Elec...

引用

AIAA Science and Technology Forum and Exposition, AIAA SciTech Forum 2025

作者： Guo, Y. Kwon, J. Sumption, M.D. Collings, E.W. Cao, D. Meng, H. Marzang, V. Liu, X. Robert, R. Layhew, G. Schafer, D. The Ohio State University College of Engineering Department of Material Science and Engineering ColumbusOH43210 United States University of Dayton School of Engineering Department of Electrical and Computer Engineering DaytonOH45469 United States Tennessee Tech University School of Engineering Department of Electrical and Computer Engineering CookevilleTN38505 United States Tennessee Tech University College of Engineering Department of Mechanical Engineering CookevilleTN38505 United States

ISBN: (数字)9781624107238

ISBN: (纸本)9781624107238

The rapid development of electric aircraft technology drives the need for highly effective cooling systems to manage the significant heat generated by high-power electronic components. In this work, liquid ammonia, with its excellent thermophysical properties including a high specific heat, and volumetric energy density, is evaluated as a good coolant (especially relevant when it will also serve aviation fuel at the same time). This work focuses on designing fuel storage and cooling strategies to address thermal management challenges both globally and locally for the electric aircraft. Aluminum was chosen as the material for the heat exchanger/heat sink due to its excellent resistance to liquid ammonia corrosion, high thermal conductivity, and low density. Using computational fluid dynamics (CFD) simulations, we analyzed the heat transfer performance and liquid ammonia flow characteristics within the aluminum heat sink. Temperature distribution of aluminum heat sink and flowing liquid are simulated to determine the viability and effectiveness of this cooling strategy. In addition, various design configurations, including fluid storage methods (cold-temperature or high-pressure) and flow channel topology, are explored to optimize the cooling efficiency. The results shows that liquid ammonia cooling is feasible to maintain the power electronics within safe operating temperatures under the given flow rates and operating conditions, especially for the cold-temperature storage strategy. This research provides valuable insights into the design and implementation of advanced cooling systems for next generation electric aircraft, indicating the potential of liquid ammonia as an effective and innovative cooling media good cooling media, despite challenges related to environmental, safety, and corrosion concerns. © 2025, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved.

关键词： Aluminum corrosion

来源：评论

学校读者我要写书评

暂无评论

An efficient fault diagnosis model using Lappet Falco optimisation based on a deep neural network for the VSI under varying load conditions

引用

International Journal of Power Electronics 2025年第3期21卷 247-278页

作者： Baste, Vaishali Shende, Dipali Idhate, Seema Deshpande, Arya Department of Electronics and Telecommunication SKNCOE Vadgaon Pune India Department of Electronics and Telecommunication PCCOER Ravet India School of Computer Engineering and Technology MIT-WPU Pune India

Numerous industrial applications employ three-phase converters that are based on insulated-gate bipolar transistors (IGBTs). However, the functioning and safety of power electronic devices and loads can be considerably impacted by IGBT faults. Maintaining high-power quality and system availability requires timely and accurate detection of power inverter failures. Constantly monitoring the failures in three-phase voltage source inverter (VSI) has greatly improved maintenance efficiency and stability. Hence, the developed research employs the discrete wavelet transform (DWT) and Lappet Falco optimised deep neural network (LFO-DNN) model to create an open circuit fault detection model for the VSI circuit. Data collection involves extracting features such as three-phase voltage, current, speed, and torque from erroneous data. The DNN classifier trained on these features uses the average three-phase current value to identify faulty switches. The VSI acting as a load with variable frequency reference is connected to a three-phase induction motor. The proposed Lappet Falco optimisation accurately yields impressive results in terms of prediction accuracy of 96.34%, precision of 96.34%, recall of 96.24%, F1 measure of 94.23%, MSE of 3.66, and specificity of 95.28%, demonstrating high efficiency for both 90% training and a k-fold value of 10. Copyright © 2025 Inderscience Enterprises Ltd.

关键词： Insulated gate bipolar transistors (IGBT)

来源：评论

学校读者我要写书评

暂无评论

A Reinforcement Learning Approach for Routing in Marine Communication Network of Fishing Vessels

引用

SN computer Science 2025年第1期6卷 1-12页

作者： Surendran, Simi Montresor, Alberto Vinodini Ramesh, Maneesha Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Amritapuri Kollam India Department of Information Engineering and Computer Science University of Trento Trento Italy Amrita Center for Wireless Networks & Applications (AmritaWNA) Amrita School of Engineering Amrita Vishwa Vidyapeetham Amritapuri Kollam India

The lack of affordable communication facilities to the shore remains a fundamental problem for fishermen engaged in deep-sea fishing. The Offshore Communication Network (OCN) is a wireless network of fishing vessels, whose goal is to provide Internet over the ocean. However, the dynamic nature of OCNs characterized by extreme weather, the difficulty of deploying additional infrastructure, wave-induced vessel movements, and high mobility causes significant challenges for traditional routing protocols. This paper proposes OCN-AR, a Q-learning-based adaptive routing strategy for ocean networks. The quality of the learning process relies on the reward function, which has been carefully designed to incorporate the most important features, including real-time forecasts of connectivity quality, path probability, link availability duration, and distance to the destination. The routing performance is evaluated through extensive simulations conducted under diverse conditions, including varying mobility scenarios, transmission rates, vessel rocking intensities, and node densities, and is compared against traditional protocols. The results demonstrate that OCN-AR significantly outperforms existing routing approaches, making it a reliable solution for maritime communication. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Fishing vessel network Marine communication network Reinforcement learning Routing protocol

来源：评论

学校读者我要写书评

暂无评论

Telugu Language Analysis with XLM-RoBERTa: Enhancing Parts of Speech Tagging for Effective Natural Language Processing

引用

SN computer Science 2025年第2期6卷 1-13页

作者： Bharathi Mohan, G. Prasanna Kumar, R. Krishna Jayanth, K. Doss, Srinath Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Chennai India Faculty of Engineering and Technology Botho University Gaborone Botswana

This research investigates the efficacy of XLM-RoBERTa, a potent deep learning architecture rooted in transformer networks, for Part-of-Speech (POS) tagging—a foundational task in Natural Language Processing (NLP). The study narrows its focus to address the formidable UCD Telugu dataset, renowned for its intricate morphological characteristics that pose challenges to accurate tagging. Through meticulous fine-tuning of the pre-trained XLM-RoBERTa model. This study adeptly discerns and exploit intricate syntactic and semantic patterns underpinning the interplay between lexical elements and their respective POS tags. The rigorous evaluation regimen, complemented by a comprehensive statistical comparison with the Multilingual BERT model, conclusively establishes XLM-RoBERTa’s superiority. This advanced model attains a remarkable accuracy score of 91%, decisively surpassing the Multilingual BERT model, which achieves an accuracy of 88%. These findings not only underscore XLM-RoBERTa’s prowess in precise POS tagging, especially in the context of morphologically rich languages like Telugu, but also carry significant implications for the broader NLP community. This research unlocks valuable insights and positions XLM-RoBERTa as an exemplar of reliability and robustness in POS tagging. Furthermore, it extends an invitation for further exploration and innovation in deep learning architectures within the NLP domain. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Bert Multilingual Bert NLP Parts-of-speech tagging Telugu XLM-RoBERTa

来源：评论

学校读者我要写书评

暂无评论

Control Structures and Algorithms for Force Feedback Bilateral Teleoperation Systems: A Comprehensive Review

引用

computer Modeling in engineering & Sciences 2025年第2期142卷 973-1019页

作者： Jiawei Tian Yu Zhou Lirong Yin Salman A.AlQahtani Minyi Tang Siyu Lu Ruiyang Wang Wenfeng Zheng Department of Computer Science and Engineering Major in Bio Artificial IntelligenceHanyang UniversityAnsan-si15577Republic of Korea Department of Geography and Anthropology Louisiana State UniversityBaton RougeLA 70803USA Department of Computer Engineering College of Computer and Information SciencesKing Saud UniversityRiyadh11574Saudi Arabia School of Automation University of Electronic Science and Technology of ChinaChengdu610054China

Force feedback bilateral teleoperation represents a pivotal advancement in control technology,finding widespread application in hazardous material transportation,perilous environments,space and deep-sea exploration,and healthcare *** paper traces the evolutionary trajectory of force feedback bilateral teleoperation from its conceptual inception to its current *** elucidates the fundamental principles underpinning interaction forces and tactile exchanges,with a specific emphasis on the crucial role of tactile *** this review,a quantitative analysis of force feedback bilateral teleoperation development trends from 2011 to 2024 has been conducted,utilizing published journal article data as the primary source of *** review accentuates classical control frameworks and algorithms,while also delving into existing research advancements and prospec-tive breakthrough ***,it explores specific practical scenarios ranging from intricate surgeries to hazardous environment exploration,underscoring the technology’s potential to revolutionize industries by augmenting human manipulation of remote *** underscores the pivotal role of force feedback bilateral teleoperation as a transformative human-machine interface,capable of shaping flexible control strategies and addressing technological *** research endeavors in force feedback bilateral teleoperation are expected to prioritize the creation of more immersive experiences,overcoming technical hurdles,fortifying human-machine collaboration,and broadening application domains,particularly within the realms of medical intervention and hazardous *** the continuous progression of technology,the integration of human intelligence and robotic capabilities is expected to produce more innovations and breakthroughs in the field of automatic control.

关键词： Force feedback bilateral teleoperation robot development history application prospect

来源：评论

学校读者我要写书评

暂无评论

A Comparison of Deep Learning and Machine Learning Methods for Water Quality Prediction 3

A Comparison of Deep Learning and Machine Learning Methods f...

引用

3rd International Conference on Intelligent Data Communication Technologies and Internet of Things, IDCIoT 2025

作者： Srinivasa Gupta, N. Valarmathi, B. Hiwanj, Ishaan Department of Manufacturing Engineering School of Mechanical Engineering Vellore Institute of Technology Tamil Nadu Vellore India Department of Software and Systems Engineering School of Computer Science Engineering and Information Systems Vellore Institute of Technology Tamil Nadu Vellore India Department of Information Technology School of Computer Science Engineering and Information Systems Vellore Institute of Technology Tamil Nadu Vellore India

ISBN: (纸本)9798331527549

Predicting water quality is essential to preserving human health and environmental sustainability. Traditional water quality assessment methods often face scalability and real-time monitoring limitations. With accuracies of 62%, 72%, 83%, 69%, 63%, 66%, 71%, 63%, and 64%, respectively, the current techniques utilized were Logistic Regression, Decision Trees, Random Forest Regressor, Extreme Gradient Boosting, Naive Bayes, K-nearest neighbors, Support Vector Machine, AdaBoost, and Bagging [9]. This study addresses these challenges by leveraging Adaptive Synthetic Sampling (ADASYN) to balance the dataset and evaluating model performance on datasets of 5,000 and 10,000 entries per class. A robust dataset obtained from Kaggle was used, with five models – Long Short-Term Memory (LSTM), Feed Forward Neural Network (FFNN), Categorical Boosting (CatBoost), Extreme Gradient Boosting (XGBoost), and Random Forest – evaluated and compared. The proposed methods demonstrate significant improvements in accuracy, with XGBoost achieving the highest accuracy of 95.53%, followed by Random Forest at 93.98%. This work underscores the importance of advanced machine learning techniques in addressing the limitations of traditional methods, enhancing accuracy, scalability, and adaptability in water quality prediction. These findings contribute to advancing environmental monitoring and management practices with reliable, data-driven insights. © 2025 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Based Enhanced Remote Power Analysis Attacks 5

Machine Learning Based Enhanced Remote Power Analysis Attack...

引用

5th International Conference on Advanced Research in Computing, ICARC 2025

作者： Kasthuripitiya, Ishan Ariyawansha, Jeewantha Ranasinghe, Tharindu Sandirigama, Manjula Jayasinghe, Darshana University of Peradeniya Department of Computer Engineering Peradeniya Sri Lanka The University of Sydney School of Electrical and Computer Engineering Sydney Australia

ISBN: (纸本)9798331530983

Remote Power Analysis (RPA) attacks have demonstrated the ability to reveal secret keys from cryptographic implementations. RPA attacks have been demonstrated on Advanced Encryption Standard (AES) circuits running on Field Programmable Gate Arrays (FPGAs). Thus far, RPA attacks were conducted using classical attacks such as correlation power analysis attacks (CPA).This paper presents a comprehensive study on the application of machine learning (ML) algorithms to improve the efficacy of RPA attacks against AES. We explore the integration of Convolutional Neural Networks (CNNs) and Multi-Layer Perceptrons (MLPs) within the state-of-the-art AISY framework (initially proposed for classical power analysis attacks) to detect and exploit RPA vulnerabilities in AES-encrypted FPGA systems. The proposed methodology involves capturing power traces using a Time-to-Digital Converter (TDC) on-chip sensor during AES operations and then using MLP and CNN models to conduct RPA attacks to evaluate security. We demonstrate that ML-enhanced RPAs can significantly increase attack success rates while requiring fewer data samples. Our results show that CNN-based RPA attacks can reveal the secret within 10,000 traces. The research underscores the need for ongoing innovation in cryptographic defences to counteract the evolving threat landscape. © 2025 IEEE.

关键词： Side channel attack

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：