检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Hou, Wenjun Cheng, Yi Xu, Kaishuai Hu, Yan Li, Wenjie Liu, Jiang Department of Computing The Hong Kong Polytechnic University Hong Kong Research Institute of Trustworthy Autonomous Systems Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China

Previous research on radiology report generation has made significant progress in terms of increasing the clinical accuracy of generated reports. In this paper, we emphasize another crucial quality that it should possess, i.e., inter-report consistency, which refers to the capability of generating consistent reports for semantically equivalent radiographs. This quality is even of greater significance than the overall report accuracy in terms of ensuring the system’s credibility, as a system prone to providing conflicting results would severely erode users’ trust. Regrettably, existing approaches struggle to maintain inter-report consistency, exhibiting biases towards common patterns and susceptibility to lesion variants. To address this issue, we propose ICON, which Improves the inter-report CONsistency of radiology report generation. Aiming to enhance the system’s ability to capture similarities in semantically equivalent lesions, our approach first involves extracting lesions from input images and examining their characteristics. Then, we introduce a lesion-aware mixup technique to ensure that the representations of the semantically equivalent lesions align with the same attributes, achieved through a linear combination during the training phase. Extensive experiments on three publicly available chest X-ray datasets verify the effectiveness of our approach, both in terms of improving the consistency and accuracy of the generated reports1 © 2024, CC BY.

关键词： Radiology

来源：评论

学校读者我要写书评

暂无评论

Geometric Reachability for Attitude control systems via Contraction Theory

arXiv

引用

arXiv 2025年

作者： Xu, Chencheng Jafarpour, Saber Zhao, Chengcheng Shi, Zhiguo Chen, Jiming College of Information Science and Electronic Engineering Zhejiang University Hangzhou China Department of Electrical Computer and Energy Engineering University of Colorado Boulder BoulderCO United States College of Control Science and Engineering Zhejiang University Hangzhou China

In this paper, we present a geometric framework for the reachability analysis of attitude control systems. We model the attitude dynamics on the product manifold SO(3) × 3 and introduce a novel parametrized family of Riemannian metrics on this space. Using contraction theory on manifolds, we establish reliable upper bounds on the Riemannian distance between nearby trajectories of the attitude control systems. By combining these trajectory bounds with numerical simulations, we provide a simulation-based algorithm to over-approximate the reachable sets of attitude systems. We show that the search for optimal metrics for distance bounds can be efficiently performed using semidefinite programming. Additionally, we introduce a practical and effective representation of these over-approximations on manifolds, enabling their integration with existing Euclidean tools and software. Numerical experiments validate the effectiveness of the proposed approach. Copyright © 2025, The Authors. All rights reserved.

关键词： Geometry

来源：评论

学校读者我要写书评

暂无评论

Enhanced Student Stress Prediction in Higher Education Using a TriBranch CNN Framework 3

Enhanced Student Stress Prediction in Higher Education Using...

引用

3rd International Conference on Integrated Circuits and Communication systems, ICICACS 2025

作者： Kunwar, Fateh Bahadur Srivastava, Satyajee Dhivya, R. Gulati, Monika Amudha, K. Raj, D. Antony Arul Geetanjali Institute of Technical Studies Department of Computer Science and Engineering Udaipur India School of Engineering and Technology Manav Rachna International Institute of Research and Studies Department of Computer Science and Engineering Faridabad India M.Kumarasamy College of Engineering Department of Information Technology Karur India Mullana Ambala India Kongunadu College of Engineering and Technology Department of Electronics and Communication Engineering Trichy India Psg College of Arts & Science Department of Software Systems Coimbatore India

ISBN: (纸本)9798331508456

Students' stress levels have a significant impact on their academic performance in higher education, thus it's important to identify signs of stress early on and intervene to improve academic performance. Machine learning has evolved into a powerful tool for stress prediction, allowing for proactive support of at-risk youngsters. In order to better anticipate college students' stress levels, this study presents an enhanced TriBranch CNN framework. In order to provide the best possible performance, the technique includes data preparation, feature extraction, and model training. The dimensionality and redundancy are reduced using Principal Component Analysis (PCA). After being trained and tested against other models, the Fully Conditional Temporal Bayesian Network (Fully CTBN) outperformed them all with an accuracy of 93.29%, beating both traditional CNN and FCN approaches. These findings demonstrate that the Fully CTBN approach is useful for identifying students who may be struggling academically and providing schools with the information they need to implement data-driven interventions that boost student well-being and performance. © 2025 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Allocation Multiplicity: Evaluating the Promises of the Rashomon Set

arXiv

引用

arXiv 2025年

作者： Jain, Shomik Wang, Margaret Creel, Kathleen Wilson, Ashia Institute for Data Systems and Society MIT United States Department of Electrical Engineering and Computer Science MIT United States Department of Philosophy and Religion Khoury College of Computer Sciences Northeastern University United States

The Rashomon set of equally-good models promises less discriminatory algorithms, reduced outcome homogenization, and fairer decisions through model ensembles or reconciliation. However, we argue from the perspective of allocation multiplicity that these promises may remain unfulfilled. When there are more qualified candidates than resources available, many different allocations of scarce resources can achieve the same utility. This space of equal-utility allocations may not be faithfully reflected by the Rashomon set, as we show in a case study of healthcare allocations. We attribute these unfulfilled promises to several factors: limitations in empirical methods for sampling from the Rashomon set, the standard practice of deterministically selecting individuals with the lowest risk, and structural biases that cause all equally-good models to view some qualified individuals as inherently risky. © 2025, CC BY.

关键词： Consensus algorithm

来源：评论

学校读者我要写书评

暂无评论

An Efficient Optimization Framework for Learning General Signed Graphs From Smooth Signals

An Efficient Optimization Framework for Learning General Sig...

引用

Sensor Array and Multichannel Signal Processing Workshop

作者： Shi-Yuk Fong Anthony Man-Cho So Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong SAR China Department of Systems Engineering and Engineering Management The Chinese University of Hong Kong Hong Kong SAR China

ISBN: (数字)9798350344813

ISBN: (纸本)9798350344820

Graph learning has been widely used in many fields to study the relationships between different entities in a dataset. We present an optimization framework based on the proximal alternating direction method of multipliers (pADMM) for learning general signed graphs from smooth signals. We show that our proposed pADMM enjoys global convergence and a local linear convergence rate. Then, we demonstrate the effectiveness of the proposed framework through numerical experiments on signed graphs. Our proposed framework provides a promising approach for learning general signed graphs from smooth signals and can be a valuable tool for data analysis and decision-making in various fields.

关键词： Data analysis Array signal processing Conferences Decision making Optimization Convergence Sensor arrays

来源：评论

学校读者我要写书评

暂无评论

On the Performance of RIS-assisted Networks with HQAM

On the Performance of RIS-assisted Networks with HQAM

引用

European Conference on Networks and Communications (EuCNC)

作者： Thrassos K. Oikonomou Dimitrios Tyrovolas Sotiris A. Tegos Panagiotis D. Diamantoulakis Panagiotis Sarigiannidis Christos Liaskos George K. Karagiannidis Department of Electrical and Computer Engineering Aristotle University of Thessaloniki Thessaloniki Greece Department of Electrical and Computer Engineering University of Western Macedonia Kozani Greece Computer Science Engineering Department University of Ioannina Ioannina Greece Artificial Intelligence & Cyber Systems Research Center Lebanese American University (LAU) Lebanon

ISBN: (数字)9798350344998

ISBN: (纸本)9798350345001

In this paper, we investigate the application of hexagonal quadrature amplitude modulation (HQAM) in reconfigurable intelligent surface (RIS)-assisted networks, specifically focusing on its efficiency in reducing the number of required reflecting elements. Specifically, we present analytical expressions for the average symbol error probability (ASEP) and propose a new metric for conditioned energy efficiency, which assesses the network’s energy consumption while ensuring the ASEP remains below a certain threshold. Additionally, we introduce an innovative detection algorithm for HQAM constellations, which demonstrates a substantial reduction in computational complexity. Finally, our study reveals that HQAM significantly enhances both the ASEP and energy efficiency compared to traditional quadrature amplitude modulation (QAM) schemes.

关键词： Measurement Wireless communication Monte Carlo methods Quadrature amplitude modulation Error probability Symbols Focusing

来源：评论

学校读者我要写书评

暂无评论

Attention Model-Driven Maddpg Algorithm for Delay and Cost Aware Placement of Service Function Chains in 5g

SSRN

引用

SSRN 2024年

作者： Munshi, Joy Sultana, Sumaya Hassan, M.D. Jahid Roy, Palash Razzaque, M.D. Abdur Alkhamees, Bader Fahad Uddin, M.D. Zia Hassan, Mohammad Mehedi Department of Computer Science and Engineering Green University of Bangladesh Narayanganj Dhaka 1461 Bangladesh Green Networking Research Group Department of Computer Science and Engineering University of Dhaka Bangladesh Department of Information Systems College of Computer and Information Sciences King Saud University Riyadh Saudi Arabia Sintef Digital Oslo Norway

The rapidly expanding applications of 5G networks necessitate strategic placement of Virtual Network Functions (VNFs) within Service Function Chains (SFCs) to minimize placement costs while delivering real-time services to users. The dual objectives of this efficient placement strategy are to simultaneously reduce resource usage costs and application service delays in the 5G network. Previous studies have limitations, typically constrained by fixed resource costs or by adopting a greedy approach for resource selection from nearby nodes. In this paper, we introduce a multi-objective linear programming (MOLP) based optimization framework designed for the placement of VNFs in SFC requests, considering a real-time pricing scheme of the resources and the demands of user applications. This framework allows for the analysis of the boundary performances regarding cost and delay, facilitating a balanced trade-off between the two. Given that this problem is proven to be NP-hard in large networks, we have also developed a Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm, which leverages an attention model-based approach for the placement of SFC VNFs. This method focuses on neighboring nodes to help agents reduce the complexity of the solution and effectively capture the dynamic nature of the network environment. Simulation experiments demonstrate that our proposed system model surpasses existing state-of-the-art approaches in terms of resource placement cost and service latency. © 2024, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

KKL Observer Synthesis for Nonlinear systems via Physics-Informed Learning

arXiv

引用

arXiv 2025年

作者： Niazi, M.B. Umar Cao, John Barreau, Matthieu Johansson, Karl Henrik Division of Decision and Control Systems Digital Futures KTH Royal Institute of Technology StockholmSE-100 44 Sweden Laboratory for Information and Decision Systems Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology CambridgeMA02139 United States Department of Electrical Engineering Stanford University StanfordCA94305 United States

This paper proposes a novel learning approach for designing Kazantzis-Kravaris/Luenberger (KKL) observers for autonomous nonlinear systems. The design of a KKL observer involves finding an injective map that transforms the system state into a higher-dimensional observer state, whose dynamics is linear and stable. The observer’s state is then mapped back to the original system coordinates via the inverse map to obtain the state estimate. However, finding this transformation and its inverse is quite challenging. We propose to sequentially approximate these maps by neural networks that are trained using physics-informed learning. We generate synthetic data for training by numerically solving the system and observer dynamics. Theoretical guarantees for the robustness of state estimation against approximation error and system uncertainties are provided. Additionally, a systematic method for optimizing observer performance through parameter selection is presented. The effectiveness of the proposed approach is demonstrated through numerical simulations on benchmark examples and its application to sensor fault detection and isolation in a network of Kuramoto oscillators using learned KKL observers. © 2025, CC BY.

关键词： State estimation

来源：评论

学校读者我要写书评

暂无评论

Spatiotemporal Traffic Analysis: Deep Hybrid Approach for Resource and Cyberattacks Prediction

Spatiotemporal Traffic Analysis: Deep Hybrid Approach for Re...

引用

International Conference on Ubiquitous Information Management and Communication (IMCOM)

作者： Hyeonmin Lee Dongchan Kim Donggyu Beom Gyurin Byun Van-Vi Vo Hyunseung Choo Department of AI Systems Engineering Sungkyunkwan University Suwon Republic of Korea Department of Electrical and Computer Engineering Sungkyunkwan University Suwon Republic of Korea Convergence Research Institute Sungkyunkwan University Suwon Republic of Korea Department of Computer Science and Engineering Sungkyunkwan University Suwon Republic of Korea

ISBN: (数字)9798331507817

ISBN: (纸本)9798331507824

This study proposes a hybrid deep learning approach to address the complexity and dynamic characteristics of modern network environments. The research integrates Graph Neural Networks (GNNs) and Convolutional Long and Short-Term Memory (ConvLSTM) networks to predict future network traffic and detect cyberattacks. The model was trained and evaluated on the CICIDS 2018 dataset which includes various types of normal and malicious traffic. Grid-based representations were created from the dataset. The attack type of each timestamp were visualized using distinct colors to the grid images. These images are sued as input of ConvLSTM model. Protocol and destination port are modeled as nodes in a graph, with volume of data flow represented as weighted edges. This graph-based representation is designed to capture unique traffic patterns of specific services and enhance the analysis of spatial and temporal features. Experimental results show a 13% increase in macro F1 score and a 4% increase in weighted F1 score, highlighting the effectiveness of the proposed method. This research highlights its potential to enable proactive responses through the prediction of potential cyberattack occurrences.

关键词： Solid modeling Protocols Image color analysis Image edge detection Telecommunication traffic Traffic control Graph neural networks Spatiotemporal phenomena Information management computer crime

来源：评论

学校读者我要写书评

暂无评论

Peer to Peer Federated Learning: Towards Decentralized Machine Learning on Edge Devices 7

Peer to Peer Federated Learning: Towards Decentralized Machi...

引用

7th South-East Europe Design Automation, computer engineering, computer Networks and Social Media Conference, SEEDA-CECNSM 2022

作者： Karras, Aristeidis Karras, Christos Giotopoulos, Konstantinos C. Tsolis, Dimitrios Oikonomou, Konstantinos Sioutas, Spyros University of Patras Decentralized Systems Computing Group Computer Engineering and Informatics Department Patras Greece University of Patras Department of Management Science and Technology Patras Greece Ionian University Department of Informatics Corfu Greece

ISBN: (数字)9798350398588

ISBN: (纸本)9798350398588

Federated Learning (FL) is an emerging technique that assures user privacy and data integrity in distributed machine learning environments. To perform so, chunks of data are trained across edge devices and a high performance cluster server maintains a local copy without exchanging it with other parties. In this work, we investigate a FL scenario in a real-world case study using 5, 10 and 20 Raspberry Pi devices acting as clients. Under this setup, we employ the widely known FedAvg algorithm which trains each client for several local epochs and then the weight of each model is aggregated. Moreover we perform experiments across imbalanced and noisy data so as to explore scalability and robustness based on real-world datasets were noise is present and we also propose two innovative algorithms where the FL scenario is considered as a peer-to-peer formulation. Ultimately, to ensure that each device is not oversampled a client-balancing Dirichlet sampling algorithm with probabilistic guarantees is proposed. © 2022 IEEE.

关键词： computer aided instruction

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：