检索结果-内蒙古大学图书馆

Mobile, Intelligent, and Ubiquitous Computing Conference (MIUCC), International

作者： Doha Thabit Bassma Alabassy M. Watheq El-Kharashi Mona Safar Mohamed Dessouky Department of Electronics Engineering and Electrical Communications Ain Shams University Cairo Egypt Department of Computer and Systems Engineering Ain Shams University Cairo Egypt

This paper introduces a high-accuracy implementation of the softmax layer in Deep Neural Networks (DNN) used in multi-category classification applications. Calculation of the exponentials or logarithms in the traditional base-e softmax is mathematically complex, which makes it hard to have a high-accuracy hardware implementation without being resource-consuming, so instead of using $e$ as the exponential base, this paper presents a hardware implementation of the base-2 softmax function that makes use of 2 as the exponential base. Thus, the complex operations in base-e softmax will be replaced by simple shift and addition operations, with simpler LUTs and higher accuracy. The implemented hardware model relies on single-precision floating-point arithmetic cores, it achieves classification accuracy equal to 100 % relative to a reference software model and has an area of 0.0802 mm 2 with a power consumption of 8.93 mW when synthesized under TSMC 28nm CMOS technology at the frequency of 1 GHz.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Responding to News Sensitively in Stock Attention Networks via Prompt-Adaptive Trimodal Model

引用

IEEE Transactions on Neural Networks and Learning systems 2025年第6期36卷 10099-10113页

作者： Haotian Liu Bowen Hu Yadong Zhou Yuxun Zhou Ministry of Education Key Laboratory for Intelligent Networks and Network Security Faculty of Electronic and Information Engineering Xi’an Jiaotong University Xi’an Shaanxi China Department of Electrical Engineering and Computer Sciences University of California at Berkeley Berkeley CA USA

Modern quantitative finance and portfolio-based investment hinge on multimedia news and historical price trends for stock movement prediction. However, prior studies overlook the long tail effect in the feature distribution of stocks, inevitably leading to biased attention and thus degrading the efficiency of utilizing news. To this end, we propose a prompt-adaptive trimodal model (PA-TMM) to overcome the biased stock attention networks and tail feature scarcity problem. In this model, sentiments automatically extracted from trimodal information serve as prompts reflecting the market’s collective mood for other entities, and the interactions among stocks are dynamically inferred for integrating both news- and price-induced movements. By leveraging the movement prompt adaptation (MPA) strategy, our model proactively adapts to the feature-imbalanced phenomenon and converges toward being responsive to the news sensitively. Extensive experiments conducted on real-world datasets consistently demonstrate not only the superiority of the proposed framework over various state-of-the-art baselines, but also its effectiveness, profitability, and robustness in Fintech. The code is accessible at https://***/lauht/PA-TMM .

关键词： Adaptation models Peer-to-peer computing Investment Heavily-tailed distribution Tail Stock markets Feature extraction Correlation Attention mechanisms Time series analysis

来源：评论

学校读者我要写书评

暂无评论

DropMix: A Textual Data Augmentation Combining Dropout with Mixup

DropMix: A Textual Data Augmentation Combining Dropout with ...

引用

2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022

作者： Kong, Fanshuang Zhang, Richong Guo, Xiaohui Mensah, Samuel Mao, Yongyi SKLSDE Beihang University Beijing China Zhongguancun Laboratory Beijing China Hangzhou Innovation Institute Beihang University Hangzhou China Department of Computer Science University of Sheffield United Kingdom School of Electrical Engineering and Computer Science University of Ottawa Canada

Overfitting is a common problem when there is insufficient data to train deep neural networks in machine learning tasks. Data augmentation regularization methods such as Dropout, Mixup, and their enhanced variants, are effective and prevalent, and achieve promising performance to overcome overfitting. However, in text learning, most of the existing regularization approaches merely adopt ideas from computer vision without considering the importance of dimensionality in natural language processing. In this paper, we argue that the property is essential to overcome overfitting in text learning. Accordingly, we present a saliency map informed textual data augmentation and regularization framework, which combines Dropout and Mixup, namely DropMix, to mitigate the overfitting problem in text learning. In addition, we design a procedure that drops and patches fine grained shapes of the saliency map under the DropMix framework to enhance regularization. Empirical studies confirm the effectiveness of the proposed approach on 12 text classification tasks. © 2022 Association for Computational Linguistics.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Studying the Impact of IBR Modeling on the Commonly Applied Transmission Line Protective Elements

Studying the Impact of IBR Modeling on the Commonly Applied ...

引用

IEEE Energy Conversion Congress and Exposition, ECCE

作者： Soham Chakraborty Paulo Henrique Pinheiro Goncalves Bainy Romulo Hangtian Lei Brian K. Johnson Scott Manson Jing Wang Rasel Mahmud Andy Hoke Cameron J. Kruse Power Systems Engineering Center National Renewable Energy Laboratory Golden Colorado USA Electrical and Computer Engineering University of Idaho Moscow Idaho USA Schweitzer Engineering Laboratories Kauai Island Utility Cooperative

ISBN: (数字)9798350376067

ISBN: (纸本)9798350376074

It is a well-accepted fact that detailed inverter-based resource (IBR) models are needed for electromagnetic transient (EMT) studies, especially for protection studies and protection designs. However, it is not clear which modeling aspects should be included in the EMT model to accurately represent the IBR fault responses. Therefore, this paper addresses this gap by evaluating the impact of key modeling aspects (DC source, inverter model, power control, current/voltage control, and current limiting) on the response of protective relay elements both qualitatively and quantitatively. Both grid-following (GFL) and grid-forming (GFM) IBR models are comprehensively evaluated through a simplified real-world transmission network by varying the IBR modeling aspects, fault locations, and types. The insights and findings from this paper will benefit the power system protection community because this work provides modeling requirements and guidelines for IBR modeling so that correct IBR models will be used for protection scheme settings design.

关键词： Radio frequency Power system protection Decision making Protective relaying Switches Predictive models Prediction algorithms Control systems Inverter-based resource Protection

来源：评论

学校读者我要写书评

暂无评论

Energy Minimization using Custom-Designed Magnetic-Spring Actuators

Energy Minimization using Custom-Designed Magnetic-Spring Ac...

引用

IEEE/RSJ International Conference on Intelligent Robots and systems (IROS)

作者： Yue Yang Fu Ali U. Kilic David J. Braun Department of Electrical and Computer Engineering SyBBURE Searle Undergraduate Research Program Vanderbilt University Advanced Robotics and Control Laboratory Advanced Robotics and Control Laboratory Vanderbilt University

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

This study introduces an innovative actuator that resembles a motor with a non-uniform permanent magnetic field. We have developed a prototype of the actuator by combining a standard motor, characterized by a uniform magnetic field, with a custom rotary magnetic spring exhibiting a non-uniform magnetic field. We have also presented a systematic computational approach to customize the magnetic field to minimize the energy consumption of the actuator when used for a user-defined oscillatory task. Experiments demonstrate that this optimized actuator significantly lowers energy consumption in a typical oscillatory task, such as pick-and-place or oscillatory limb motion during locomotion, compared to conventional motors. Our findings imply that incorporating task-optimized non-uniform permanent magnetic fields into conventional motors and direct-drive actuators could enhance the energy efficiency of robotic systems.

关键词： Actuators Energy consumption Limbs Systematics Prototypes Motors Minimization Magnetic fields Springs Standards

来源：评论

学校读者我要写书评

暂无评论

Development of A Surveillance System to Detect Forest Fire and Smoke Using Deep Neural Networks 3

Development of A Surveillance System to Detect Forest Fire a...

引用

3rd IEEE KhPI Week on Advanced Technology, KhPI Week 2022

作者： Subramanian, Malliga Shanmugavadivel, Kogilavani Muthusamy, Suresh Bajaj, Mohit Rubanenko, Olena Danylchenko, Dmytro Kongu Engineering College Department of Computer Science and Engineering Erode India Department of Electronics and Communication Engineering Tamil Nadu Erode India Department of Electrical Engineering Dehradun248002 India Vinnytsia National Technical University Department of Electrical Stations and Systems Vinnytsya Ukraine National Technical University «kharkiv Polytechnic Institute» Department of Electric Power Transmission Kharkiv Ukraine

ISBN: (数字)9798350399202

ISBN: (纸本)9798350399202

Forest fires are a major environmental hazard, causing economic, ecological, and human harm. Forest fires are natural flames that occur in forests, shrubs, and grasslands. Wildfires are usually caused by lightning or human error and occur during dry hot summers. Predicting such an environmental issue is crucial to reducing the threat. WSN is one of the innovative technologies and approaches developed to anticipate and detect forest fires. The accuracy of such fire detection technologies ranges from 80% to 91%. In this work, deep learning identification systems based on Convolutional Neural Networks (CNN) namely VGG16 and Mobile Net via Teachable Machine are presented to automatically detect fire in its early phases. Fire, No Fire, Smoke, and Smoke Fire are the four classes of images of forest fires identified in this work. With 98.5% accuracy on the Fire Image dataset, Mobile Net via Teachable machine outperforms the literature when it comes to predicting the fire. © 2022 IEEE.

关键词： Fires

来源：评论

学校读者我要写书评

暂无评论

Worst-Case Integrity Attacks and Resilient State Estimation With Partially Secured Measurements

引用

IEEE Transactions on Control of Network systems 2025年

作者： Zhou, Jing Shang, Jun Chen, Tongwen University of Alberta Department of Electrical and Computer Engineering EdmontonABT6G 1H9 Canada Tongji University Department of Control Science and Engineering Shanghai Institute of Intelligent Science and Technology National Key Laboratory of Autonomous Intelligent Unmanned Systems Frontiers Science Center for Intelligent Autonomous Systems Shanghai200092 China

This paper examines the problem of optimal deception attacks against state estimation with partially secured measurements, where smart sensors transmit innovation sequences to the remote end for information fusion. Due to resource limitations or defensive countermeasures, the adversary can only modify data packets transmitted through unreliable channels. Meanwhile, the attack should be synthesized with sophistication to deceive an anomaly detector. To investigate the vulnerabilities of such estimation systems without feedback links and enhance security performance, the optimal attack policy is derived by formulating and explicitly solving a convex optimization problem, with the goal of maximizing the sum of estimation errors. Subsequently, a novel attack detection and resilient state estimation algorithm is proposed to ensure an acceptable level of estimation accuracy. The theoretical performance metrics, including false alarm rates for the proposed detector, are provided. Finally, the effectiveness of the results is confirmed through numerical examples. © 2014 IEEE.

关键词： Optimal detection

来源：评论

学校读者我要写书评

暂无评论

Elucidation of the Relationship Between a Song's Spotify Descriptive Metrics and its Popularity on Various Platforms

Elucidation of the Relationship Between a Song's Spotify Des...

引用

IEEE Annual International computer Software and Applications Conference (COMPSAC)

作者： Laura Colley Andrew Dybka Adam Gauthier Jacob Laboissonniere Alexandre Mougeot Nayeeb Mowla Kevin Dick Hoda Khalil Gabriel Wainer Department of Systems&#x0026 Computer Engineering Department of Electrical Engineering Carleton University Ottawa Ontario

ISBN: (数字)9781665488105

ISBN: (纸本)9781665488112

The music industry and personal music consumption have evolved dramatically with the advent of streaming plat-forms. In this evolving landscape, there is considerable interest in understanding what factors contribute to a song's popularity. Extrinsic (i.e. non-acoustic) features of a given song, such as the record label, and/or intrinsic (i.e. acoustic) features such as its energy may contribute to popularity on a given digital platform. In this work, we, for the first time, sought to systematically study how a song's Spotify acoustic descriptive features correlated with popularity metrics on various Internet platforms. Since each platform defines “popularity” according to platform-specific metrics, a large-scale correlation-based analysis was generated. The digital platforms considered in this article are Google Trends, WhoSampled, TikTok, Twitter, YouTube, and the Billboard Top-100. Platform-specific scrapers were created and all data was aggregated with the Spotify Echo Nest dataset of descriptive acoustic metrics. While the majority of correlations were unre-markable considering both Spearman and Pearson coefficients, a number of corroborating and contradictory findings resulted, with notable implications for acoustic features on various digital platforms. Notably, the YouTube view count was found to be positively correlated to the Spotify song popularity (p = 0.822), year (p = 0.600), and energy (p = 0.455) and moderately negatively correlated to accousticness (p = −0.542) and instrumentalness (p = −0.345). All reproducing code and aggregated data from this work are open-source for use by the broader research community.

关键词： Measurement Correlation Video on demand Codes Systematics Social networking (online) Blogs

来源：评论

学校读者我要写书评

暂无评论

Control of Isolated AC Microgrids with Constant Power Loads: A Set Invariance Approach

Control of Isolated AC Microgrids with Constant Power Loads:...

引用

Mediterranean Conference on Control and Automation (MED)

作者： Grigoris Michos George C. Konstantopoulos Paul A. Trodden Visakan Kadirkamanathan Department of Automatic Control & Systems Engineering University of Sheffield Mappin Street Sheffield UK Department of Electrical and Computer Engineering University of Patras Rion Greece

This paper proposes a robust control scheme for isolated AC Microgrids, where each node is connected locally to a constant power load (CPL). Contrary to many approaches in the literature, we consider the explicit model of the inverter dynamics and separate the overall system into two parts; a nominal subsystem parametrized by a nominal load and an error subsystem describing the difference between the true and the nominal voltage, resulting from perturbations of the load demand. In the presented analysis, we investigate the non-linear structure of the CPL in order to analytically describe its geometric effect on the network dynamics. We exploit this information to propose mild conditions on the tuning parameters such that a positive invariant set for the error dynamics exists and the distance between the true and the nominal voltage trajectories is bounded at all times. We demonstrate the properties of the proposed control scheme in a simulated scenario.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Taxonomy of Defenses against Memory Corruption Attacks 44

A Taxonomy of Defenses against Memory Corruption Attacks

引用

44th International Convention on Information, Communication and Electronic Technology, MIPRO 2021

作者： Novkovic, Bojan University of Zagreb Faculty of Electrical Engineering and Computing Department of Electronics Microelectronics Computer and Intelligent Systems Zagreb Croatia

ISBN: (纸本)9789532331011

Vulnerabilities caused by memory corruption related bugs are a pervasive threat, continually undermining the security of the whole computing environment. The lack of memory safety mechanisms in indispensable systems programming languages like C or C++ leaves plenty of room for programmer-induced errors which often result in catastrophic security breaches. We analyze the root causes of these vulnerabilities, providing a concise overview of memory corruption bugs and related attacks. Although present in a number of programming languages, we use the C/C++ programming language as a basis for our analyses. We categorize existing defensive mechanisms based on the attack techniques they focused on preventing and give a brief insight into state-of-the-art defenses introduced throughout the years, with a special focus on operating system defenses. © 2021 Croatian Society MIPRO.

关键词： C++ (programming language)

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：