检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Alehdaghi, Mahdi Bhattacharya, Rajarshi Shamsolmoali, Pourya Cruz, Rafael M.O. Heritier, Maguelonne Granger, Eric LIVIA ILLS Dept. of Systems Engineering École De Technologie Supérieure MontrealQC Canada Dept. of Computer Science University of York York United Kingdom Genetec Inc MontrealQC Canada

Deep learning has provided considerable advancements for multimedia systems, yet the interpretability of deep models remains a challenge. State-of-the-art post-hoc explainability methods, such as GradCAM, provide visual interpretation based on heatmaps but lack conceptual clarity. Prototype-based approaches, like ProtoPNet and PIPNet, offer a more structured explanation but rely on fixed patches, limiting their robustness and semantic consistency. To address these limitations, a part-prototypical concept mining network (PCMNet) is proposed that dynamically learns interpretable prototypes from meaningful regions. PCMNet clusters prototypes into concept groups, creating semantically grounded explanations without requiring additional annotations. Through a joint process of unsupervised part discovery and concept activation vector extraction, PCMNet effectively captures discriminative concepts and makes interpretable classification decisions. Our extensive experiments comparing PCMNet against state-of-the-art methods on multiple datasets show that it can provide a high level of interpretability, stability, and robustness under clean and occluded scenarios. © 2025, CC BY.

关键词： Interactive computer systems

来源：评论

学校读者我要写书评

暂无评论

Uncertainty Propagation in Stochastic systems via Mixture Models with Error Quantification

arXiv

引用

arXiv 2024年

作者： Figueiredo, Eduardo Patane, Andrea Lahijanian, Morteza Laurenti, Luca Delft Center for Systems and Control TU Delft Netherlands Dept. of Aerospace Eng. Sciences and Computer Science CU Boulder United States School of Computer Science and Statistics Trinity College Dublin Ireland

Uncertainty propagation in non-linear dynamical systems has become a key problem in various fields including control theory and machine learning. In this work, we focus on discrete-time non-linear stochastic dynamical systems. We present a novel approach to approximate the distribution of the system over a given finite time horizon with a mixture of distributions. The key novelty of our approach is that it not only provides tractable approximations for the distribution of a nonlinear stochastic system but also comes with formal guarantees of correctness. In particular, we consider the Total Variation (TV) distance to quantify the distance between two distributions and derive an upper bound on the TV between the distribution of the original system and the approximating mixture distribution derived from our framework. We show that in various cases of interest, including in the case of Gaussian noise, the resulting bound can be efficiently computed in closed form. This allows us to quantify the correctness of the approximation and to optimize the parameters of the resulting mixture distribution to minimize such distance. The effectiveness of our approach is illustrated on several benchmarks from the control community. © 2024, CC BY.

关键词： Stochastic models

来源：评论

学校读者我要写书评

暂无评论

Hyperbolic DoA Estimation and Static Localization: Comparative Study of Bluetooth and RFID

Hyperbolic DoA Estimation and Static Localization: Comparati...

引用

IEEE International Conference on RFID

作者： Georgios Andreadis Evangelos Giannelos Panos N. Alevizos Aggelos Bletsas School of Electrical and Computer Engineering Technical University of Crete Chania Greece Adveos Microelectronic Systems Athens Greece Dept. of Electrical and Computer Engineering WINLAB Rutgers University The State Univ. of New Jersey

ISBN: (数字)9798350373592

ISBN: (纸本)9798350373608

This work presents a comparison between RFID and BLE regarding tag localization accuracy under static conditions, where both the antennas and the tag remain stationary. The comparison utilizes a hyperbolic localization technique based on phase differences, which introduces linear approximations of hyperbolas to calculate 3D direction of arrival (DoA) and estimate the tag's 3D position. However, this technique is highly sensitive to multipath noise. The spatial and frequency diversity in BLE aids in mitigating multipath issues, resulting in accurate estimations. In contrast, RFID lacks compatibility with such diversity. This comparison motivates the exploration of developing new algorithms that ensure consistent results for both technologies. Additionally, it suggests the possibility of merging BLE and RFID technologies into a single tag, minimizing energy consumption while also providing accurate localization estimates in static scenarios.

关键词： Location awareness Accuracy Direction-of-arrival estimation Three-dimensional displays Noise Merging Estimation

来源：评论

学校读者我要写书评

暂无评论

Development of Augmented Reality-based Dental Surgical Guide System for Clinical Use

Development of Augmented Reality-based Dental Surgical Guide...

引用

IEEE Global Conference on Consumer Electronics (GCCE)

作者： Yusuke Fukuda Akira Takebayashi Hiroaki Yamamoto Toru Kano Dept. Major in Computer and Information Sience Ibaraki University Ibaraki Japan Medical Engineering Lab. Saitama Japan Tsuda Dental Clinic Ibaraki Japan Dept. of Mechanical Systems Engineering Ibaraki University Ibaraki Japan

ISBN: (数字)9798350355079

ISBN: (纸本)9798350355086

Surgical guide plates, which are used in dental implant treatments, have some limitations that cause problems for the operator and patient. This study aims to develop an augmented-reality-based surgical guide system to solve these problems and discusses its use in clinical practice. Our surgical guide system achieved a good accuracy in drilling simulations on training typodonts, demonstrating its potential in clinical practice.

关键词： Training Drilling Electric potential Accuracy Surgery Implants Dentistry Consumer electronics

来源：评论

学校读者我要写书评

暂无评论

Simulation Model of the Hydraulic Drive of the Rotary Column of the Robot Manipulator 5

Simulation Model of the Hydraulic Drive of the Rotary Column...

引用

5th International Conference on control systems, Mathematical Modeling, Automation and Energy Efficiency, SUMMA 2023

作者： Kolpashchikov, Sergey A Terekhin, Michail A Kozlov, Evgeny V Portnov, Artem A Kalugin, Dmitry S Surguchev, Ivan V Samara State Technical University Dept. Automation and Control in Technical Systems Samara Russia Samara State University of Railway Transport Higher School of Medical Engineering Dept. Digital Technologies Institute of Innovative Development FSBEI HE SamSMU MOH Russia Samara Russia State University of Railway Transport Dept. Digital Technologies Samara Samara Russia State University of Railway Transport Institute of Innovative Development FSBEI HE SamSMU MOH Russia Dept. Digital Technologies Samara Samara Russia

ISBN: (纸本)9798350309126

The purpose of the work is to develop a simulation model of the hydraulic drive of the rotary column of the robot manipulator, which provides an adequate similarity of the proposed model with a real object. Modeling of the system was carried out with the help of modern FluidSIM Hydraulics software, which allows to intuitively reproduce and evaluate the operation of the system without using special complex mathematical structures for describing the research model. The result modeling is the transient processes hydraulic motor rotation, position value hydraulic distributor, pressure values in hydraulic lines, pushbutton state. The obtained dependences of numerical values at various time intervals, presented in the corresponding section of this work, allow us to conclude that the developed model adequately describes the nature and behavior of the object under study, corresponding to the normal mode of operation of the hydraulic system. When changing the initial data of the simulation model, it is possible to achieve other operating modes, for example, emergency ones. The developed system rotates clockwise and counterclockwise, control is carried out using a relay-contactor system. The simulation of the system was carried out taking into account the compensation of hydraulic shocks in the hydraulic lines when changing the direction of rotation in the opposite direction. Based on the results obtained, it can be concluded that the efficiency of the hydraulic drive of the rotary column of the robot manipulator depends on several main factors, such as speed, load and pressure in the hydraulic lines. Changing these parameters can affect the system both positively, leading to an increase in the efficiency of the system as a whole, and negatively and lead to a decrease in positioning accuracy, an increase in the reaction time of the system and the efficiency of power transmission. The results obtained are of interest not only for engineers of hydraulic drive systems, but also

关键词： Machine tools

来源：评论

学校读者我要写书评

暂无评论

Load Demand Prediction for Electric Vehicles Smart Charging through Consensus-based Federated Learning

Load Demand Prediction for Electric Vehicles Smart Charging ...

引用

Mediterranean Conference on control and Automation (MED)

作者： Danilo Menegatti Antonio Pietrabissa Sabato Manfredi Alessandro Giuseppi Dept. of Computer Control and Management Engineering (DIAG) University of Rome ”La Sapienza” Rome Italy Dept. of Electrical Engineering and Information Technology University of Naples ”Federico II” Naples Italy

Having access to a reliable and accurate prediction of the short-term power demand is a fundamental step for the widespread adoption of Electric Vehicles (EVs), as their charges may have a significant impact on the power system balancing. In this direction, we propose a short-term load demand predictor, based on distributed Long Short-Term Memory Networks, that employs consensus and fully-decentralized Federated Learning (FL) algorithms to seek cooperation among multiple points of charge without the requirement of sharing any user-related data.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Iterative Robust Experiment Design for MIMO System Identification via the S-Lemma

Iterative Robust Experiment Design for MIMO System Identific...

引用

control Technology and Applications (CCTA),

作者： Nic Dirkx Koen Tiels Tom Oomen ASML Veldhoven The Netherlands Dept. of Mechanical Engineering Control Systems Technology Eindhoven University of Technology Eindhoven The Netherlands Department of Mechanical Engineering Control Systems Technology Section Eindhoven University of Technology Eindhoven The Netherlands Delft Center for Systems and Control Delft University of Technology Delft The Netherlands

Optimal input design plays an important role in system identification for complex and multivariable systems. A known paradox in input design is that the optimal inputs depend on the true but unknown system. The aim of this paper is to design inputs for multivariable systems that are robust to all system variations within a given continuous uncertainty set. In the presented approach, the robust design problem is cast as an infinite-dimensional min-max optimization problem, and tackled via the S-lemma in an iterative approximation scheme. Experimental results from a multivariable motion system show that the algorithm enables significant robustness improvements.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Exploring User Experience and Challenges with Mobile Wallet Adoption in Emerging Markets

Exploring User Experience and Challenges with Mobile Wallet ...

引用

Technology Innovation and Its Applications (ICTIIA), International Conference on

作者： Priandito Ninda Lutfiani Natasya Aprila Yusuf Thomas Sumarsan Goh Muchlisina Madani Dept. of Informatics Engineering University of Raharja Tangerang Indonesia Dept. of Digital Business University of Raharja Tangerang Indonesia Dept. of Information Systems University of Raharja Tangerang Indonesia Dept. of Management Universitas Methodist Indonesia Medan Indonesia Dept. of Computer System University of Raharja Tangerang Indonesia

ISBN: (数字)9798350351613

ISBN: (纸本)9798350351620

This study explores user experiences and challenges in mobile wallet adoption in emerging markets using the UTAUT model. These markets face rapid technological growth alongside infrastructural and economic challenges. The research examines how factors like performance expectancy, effort expectancy, social influence, and facilitating conditions influence adoption. Data from 80 mobile wallet users was collected through an online survey and analyzed using SmartPLS 4.1.0.6. The results show that performance and effort expectancy predict behavioral intention, while facilitating conditions affect actual usage. The study highlights the role of infrastructure in supporting adoption and validates the UTAUT model's relevance in less developed regions. These findings provide insights into enhancing financial inclusion in emerging markets.

关键词： Surveys Technological innovation Soft sensors Decision making User interfaces User experience Security Sustainable development Socioeconomics Faces

来源：评论

学校读者我要写书评

暂无评论

Polarity Classification of Twitter Data Using Machine Learning Approach

Polarity Classification of Twitter Data Using Machine Learni...

引用

Human-Centered Cognitive systems (HCCS)

作者： Waqas Haider Bangyal Muddesar Iqbal Ahsan Bashir George Ubakanma Dept. of Computer Science Kohsar Univeristy Murree Muree Pakistan Cognitive Systems Research Centre School of Engineering School of Engineering London South Bank University Dept. of Computer Science Abasyn University Islamabad Paksitan

Sentiment analysis has become very important nowadays because there are lots of social media platforms peoples are using to express their opinion. Twitter is one of the most popular social media platforms which is used for microblogs. People use to express their opinion on current affairs, and there is a challenge for researchers to classify the sentiment accurately. In this research study, we proposed a greatly efficient technique for the detection of fake news on covid-19. The data set of fake news is taken from the corpus and executes the NLP cycle. In this research, we applied five machine learning to predict the sentiment of fake or real news. Support Vector Machine, Logistic Regression, KNN, Decision Trees, and Random Forest are machine learning classifiers used in this research, and results are compared.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Novel Decentralized Coordination control Scheme for the Complex Transactive Energy Prosumers

A Novel Decentralized Coordination Control Scheme for the Co...

引用

Energy, Power, Environment, control, and Computing (ICEPECC), International Conference on

作者： Muhammad Ikram Zakir Ullah Muhammad Nasir Salman Ahmed Safdar Nawaz Khan Marwat Dept. of Computer Systems Engineering University of Engineering and Technology Peshawar Pakistan Dept. of Electrical Power Engineering University of Engineering and Technology Peshawar Pakistan Deptment of Physics Allama Iqbal Open University Islamabad Pakistan

The significant penetration of renewable energy sources (RES) in the smart grid (SG) provides a new landscape for researchers to develop an optimal energy management model. The households with RES become prosumers to provide the surplus energy in the local community. The coordination and control of such RES prosumers are important to estimate the available energy in the system and the total demand required. In this paper, we proposed a decentralized coordination control approach for transactive energy systems (TES). The proposed TES is capable of coordinating and controlling a complex network of prosumers with limited control information. The consensusability and graph-theoretic schemes are used for prosumer RES nodes. In the proposed model, energy trading among large-scale prosumers has been demonstrated. The algorithm is capable to provide the total surplus energy available across the network and total demand in an autonomous fashion. The coordination control among prosumers is achieved through a computationally efficient approach without sharing the prosumer’s profile to preserve the security and privacy of the prosumers. To validate the proposed scheme, we perform extensive simulations on decentralized 450 prosumers. The resiliency of the proposed scheme is confirmed through dynamic and contingent communication topologies for energy trading ratios, cumulative returns, and convergence ratios. The cumulative returns improved from 0.447% to 24.8% while the energy trade among the prosumers recorded from -0.89 to 2.09. The final return is enhanced from −1.95% to 22.3% due to internal and external coordination control of nonlinear loads.

关键词：

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：