检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Wang, Guanghui Acharya, Krishna Lakshmikanthan, Lokranjan Muthukumar, Vidya Ziani, Juba College of Computing Georgia Institute of Technology United States School of Industrial and Systems Engineering Georgia Institute of Technology United States School of Electrical and Computer Engineering Georgia Institute of Technology United States

Performative prediction models account for feedback loops in decision-making processes where predictions influence future data distributions. While existing work largely assumes insensitivity of data distributions to small strategy changes, this assumption usually fails in real-world competitive (i.e. multi-agent) settings. For example, in Bertrand-type competitions, a small reduction in one firm’s price can lead that firm to capture the entire demand, while all others sharply lose all of their customers. We study a representative setting of multi-agent performative prediction in which insensitivity assumptions do not hold, and investigate the convergence of natural dynamics. To do so, we focus on a specific game that we call the "Bank Game", where two lenders compete over interest rates and credit score thresholds. Consumers act similarly as to in a Bertrand Competition, with each consumer selecting the firm with the lowest interest rate that they are eligible for based on the firms’ credit thresholds. Our analysis characterizes the equilibria of this game and demonstrates that when both firms use a common and natural no-regret learning dynamic—exponential weights—with proper initialization, the dynamics always converge to stable outcomes despite the general-sum structure. Notably, our setting admits multiple stable equilibria, with convergence dependent on initial conditions. We also provide theoretical convergence results in the stochastic case when the utility matrix is not fully known, but each learner can observe sufficiently many samples of consumers at each time step to estimate it, showing robustness to slight mis-specifications. Finally, we provide experimental results that validate our theoretical findings. © 2025, CC BY.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

GACL-Net:Hybrid Deep Learning Framework for Accurate Motor Imagery Classification in Stroke Rehabilitation

引用

computers, Materials & Continua 2025年第4期83卷 517-536页

作者： Chayut Bunterngchit Laith H.Baniata Mohammad H.Baniata Ashraf ALDabbas Mohannad A.Khair Thanaphon Chearanai Sangwoo Kang State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of AutomationChinese Academy of SciencesBeijing100190China School of Computing Gachon UniversitySeongnam13120Republic of Korea Computer Science Department Faculty of Information TechnologyThe World Islamic Sciences and Education UniversityAmman11947Jordan Intelligent Systems Department Faculty of Artificial IntelligenceAl-Balqa Applied UniversityAl-Salt19117Jordan IT Infrastructure Department Qatrana Cement CompanyAmman11821Jordan Division of Industrial and Logistics Engineering Technology Faculty of Engineering and TechnologyKing Mongkut’s University of Technology North BangkokRayong CampusRayong21120Thailand

Stroke is a leading cause of death and disability worldwide,significantly impairing motor and cognitive *** rehabilitation is often hindered by the heterogeneity of stroke lesions,variability in recovery patterns,and the complexity of electroencephalography(EEG)signals,which are often contaminated by *** classification of motor imagery(MI)tasks,involving the mental simulation of movements,is crucial for assessing rehabilitation strategies but is challenged by overlapping neural signatures and patient-specific *** address these challenges,this study introduces a graph-attentive convolutional long short-term memory(LSTM)network(GACL-Net),a novel hybrid deep learning model designed to improve MI classification accuracy and ***-Net incorporates multi-scale convolutional blocks for spatial feature extraction,attention fusion layers for adaptive feature prioritization,graph convolutional layers to model inter-channel dependencies,and bidi-rectional LSTM layers with attention to capture temporal *** on an open-source EEG dataset of 50 acute stroke patients performing left and right MI tasks,GACL-Net achieved 99.52%classification accuracy and 97.43%generalization accuracy under leave-one-subject-out cross-validation,outperforming existing state-of-the-art ***,its real-time processing capability,with prediction times of 33–56 ms on a T4 GPU,underscores its clinical potential for real-time neurofeedback and adaptive *** findings highlight the model’s potential for clinical applications in assessing rehabilitation effectiveness and optimizing therapy plans through precise MI classification.

关键词： Motor imagery EEG stroke rehabilitation deep learning brain-computer interface

来源：评论

学校读者我要写书评

暂无评论

Long baseline underwater source localization based on deep K-Means++ clustering in complex underwater environments

引用

Digital Signal Processing: A Review Journal 2025年 164卷

作者： Dai, Yawen Yang, Lei Cao, Yifei Reconfigurable and Intelligent Computing Laboratory School of Computer and Information Anhui Polytechnic University Wuhu241000 China School of Network Engineering Wuhu Institute of Technology Wuhu241000 China School of Electrical Engineering Anhui Polytechnic University Wuhu241000 China

Long baseline localization relies on trilateration, with the least squares method being utilized to determine the unique position of underwater sources. However, it is highly sensitive to distance measurements from any of the reference beacons to the source. Furthermore, achieving continuous and stable high-precision distance measurements presents a significant challenge in complex marine environments. To address this practical problem, this paper proposes a long baseline underwater source localization method based on deep K-Means++ clustering. A three-layer stack denoising autoencoder (SDA) was utilized to extract the features of the trilateration results. Subsequently, the K-Means++ algorithm was utilized to conduct multi-source location cluster analysis and fine-tuning. The experimental results demonstrate that, compared to the existing Kmeans-based long baseline localization method, the proposed approach demonstrates a modest enhancement in localization accuracy. Concurrently, it has led to an average increase of 7 percentage points in normalized mutual information (NMI), a rise of 8 percentage points in the adjusted Rand index (ARI), and an improvement of 2.5 percentage points in clustering accuracy (ACC). This not only ensures the stability and robustness of accuracy against ocean noise in long baseline mode but also enhances the operational efficiency and versatility of clustering methods applied within this field. © 2025 Elsevier Inc.

关键词： Least squares approximations

来源：评论

学校读者我要写书评

暂无评论

A Training-Free Length Extrapolation Approach for LLMs: Greedy Attention Logit Interpolation (GALI)

arXiv

引用

arXiv 2025年

作者： Li, Yan Zhang, Tianyi Li, Zechuan Han, Soyeon Caren School of Computer Science The University of Sydney Sydney Australia School of Computing and Information Systems The University of Melbourne Melbourne Australia College of Electrical and Information Engineering Hunan University Changsha China

Transformer-based Large Language Models (LLMs) struggle to process inputs exceeding their training context window, with performance degrading due to positional out-of-distribution (O.O.D.) that disrupt attention computations. Existing solutions, fine-tuning and training-free methods, are limited by computational inefficiency, attention logit outliers or loss of local positional information. To address this, we propose Greedy Attention Logit Interpolation (GALI), a training-free length extrapolation method that maximizes the utilization of pretrained positional intervals while avoiding attention logit outliers through attention logit interpolation. The result demonstrates that GALI consistently outperforms state-of-the-art training-free methods. Our findings reveal that LLMs interpret positional intervals unevenly within their training context window, suggesting that extrapolating within a smaller positional interval range yields superior results—even for short-context tasks. GALI represents a significant step toward resolving the positional O.O.D. challenge, enabling more reliable long-text understanding in LLMs. Our implementation of GALI, along with the experiments from our paper, is open-sourced at https://***/AcademyCityL/GALI. © 2025, CC BY.

关键词： Extrapolation

来源：评论

学校读者我要写书评

暂无评论

Accurate, provable, and fast nonlinear tomographic reconstruction: A variational inequality approach

arXiv

引用

arXiv 2025年

作者： Lou, Mengqi Verchand, Kabir Aladin Fridovich-Keil, Sara Pananjady, Ashwin Georgia Institute of Technology School of Industrial and Systems Engineering United States School of Electrical and Computer Engineering United States University of Cambridge Statistical Laboratory United Kingdom Stanford University Department of Electrical Engineering United States

We consider the problem of signal reconstruction for computed tomography (CT) under a nonlinear forward model that accounts for exponential signal attenuation, a polychromatic X-ray source, general measurement noise (e.g. Poisson shot noise), and observations acquired over multiple wavelength windows. We develop a simple iterative algorithm for single-material reconstruction, which we call EXACT (EXtragradient Algorithm for Computed Tomography), based on formulating our estimate as the fixed point of a monotone variational inequality. We prove guarantees on the statistical and computational performance of EXACT under practical assumptions on the measurement process. We also consider a recently introduced variant of this model with Gaussian measurements, and present sample and iteration complexity bounds for EXACT that improve upon those of existing algorithms. We apply our EXACT algorithm to a CT phantom image recovery task and show that it often requires fewer X-ray projection exposures, lower source intensity, and less computation time to achieve similar reconstruction quality to existing methods. © 2025, CC BY.

关键词： computerized tomography

来源：评论

学校读者我要写书评

暂无评论

Raising Awareness of Location Information Vulnerabilities in Social Media Photos using LLMs 25

Raising Awareness of Location Information Vulnerabilities in...

引用

Proceedings of the 2025 CHI Conference on Human Factors in computing systems

作者： Ying Ma Shiquan Zhang Dongju Yang Zhanna Sarsenbayeva Jarrod Knibbe Jorge Goncalves School of Computing and Information Systems University of Melbourne Melbourne Victoria Australia School of Computing and Information System University of Melbourne Melbourne Victoria Australia School of Computer Science University of Sydney Sydney Australia School of Electrical Engineering and Computer Science The University of Queensland St Lucia QLD Australia

来源：评论

学校读者我要写书评

暂无评论

Multi-Vehicle Cooperative Persistent Coverage for Random Target Search

引用

IEEE Robotics and Automation Letters 2025年第7期10卷 6680-6687页

作者： Li, Zhuo Li, Guangzheng Sadeghi, Alireza Sun, Jian Wang, Gang Wang, Jialin Beijing Institute of Technology National Key Lab of Autonomous Intelligent Unmanned Systems School of Automation Beijing100084 China Beijing Institute of Technology Chongqing Innovation Center Chongqing401120 China University of Minnesota Department of Electrical and Computer Engineering MinneapolisMN55455 United States Intelligent Game and Decision Laboratory China Academy of Launch Vehicle Technology Beijing100071 China

This letter investigates the target search problem for a network of autonomous vehicles, aiming to maximize the detection of randomly appearing targets within a given area. Considering no prior knowledge of the targets is available, we propose a multi-vehicle cooperative persistent coverage scheme under the framework of multi-agent reinforcement learning, in contrast to heuristic and model-based optimization methods in existing works. We model the persistent coverage problem as a partially observable Markov decision process (POMDP) due to the vehicles' limited observation ranges, and introduce a knowability map to characterize their knowledge of the target area. Each vehicle employs a distributed estimator, leveraging its own observations and shared information from neighboring vehicles, to construct a globally estimated knowability map—thereby mitigating partial observability. The persistent coverage policies are learned with the architecture of centralized training and distributed execution, enabling cooperative and efficient target search by fully exploiting shared information. Moreover, we propose an adaptive partition method for the target area to ensure a fixed dimension of the state space in the POMDP, which can improve scalability of the learned policy to target areas with various sizes. Simulations validate effectiveness and scalability of the proposed cooperative scheme. © 2016 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Unknown-Aware Bilateral Dependency Optimization for Defending Against Model Inversion Attacks

引用

IEEE Transactions on Pattern Analysis and Machine Intelligence 2025年 PP卷 PP页

作者： Peng, Xiong Liu, Feng Wang, Nannan Lan, Long Liu, Tongliang Cheung, Yiu-Ming Han, Bo Hong Kong Baptist University Department of Computer Science Hong Kong The University of Melbourne School of Computing and Information Systems Australia Xidian University State Key Laboratory of Integrated Services Networks School of Telecommunications Engineering Shaanxi Xi'an710071 China National University of Defense Technology State Key Laboratory of High Performance Computing School of Computer Hunan Changsha410000 China The University of Sydney Sydney AI Centre School of Computer Science Faculty of Engineering Australia

By abusing access to a well-trained classifier, model inversion (MI) attacks pose a significant threat as they can recover the original training data, leading to privacy leakage. Previous studies mitigated MI attacks by imposing regularization to reduce the dependency between input features and outputs during classifier training, a strategy known as unilateral dependency optimization. However, this strategy contradicts the objective of minimizing the supervised classification loss, which inherently seeks to maximize the dependency between input features and outputs. Consequently, there is a trade-off between improving the model's robustness against MI attacks and maintaining its classification performance. To address this issue, we propose the bilateral dependency optimization strategy (BiDO), a dual-objective approach that minimizes the dependency between input features and latent representations, while simultaneously maximizing the dependency between latent representations and labels. BiDO is remarkable for its privacy-preserving capabilities. However, models trained with BiDO exhibit diminished capabilities in out-of-distribution (OOD) detection compared to models trained with standard classification supervision. Given the open-world nature of deep learning systems, this limitation could lead to significant security risks, as encountering OOD inputs—whose label spaces do not overlap with the in-distribution (ID) data used during training-is inevitable. To address this, we leverage readily available auxiliary OOD data to enhance the OOD detection performance of models trained with BiDO. This leads to the introduction of an upgraded framework, unknown-aware BiDO (BiDO+), which mitigates both privacy and security concerns. As a highlight, with comparable model utility, BiDO-HSIC+ reduces the FPR95 by 55.02% and enhances the AUCROC by 9.52% compared to BiDO-HSIC, while also providing superior MI robustness. © 1979-2012 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

Theoretical Characterization of the Electro-optical Properties of New Molecules for OLED Devices

Theoretical Characterization of the Electro-optical Properti...

引用

Innovative Research in Applied Science, engineering and Technology (IRASET),Conference on

作者： Imane El Mhamedi Zakariae El Madani Abdelhafid Yahya Zakaria El Malki Materials and Systems Control (MMSC) Computer Engineering and Intelligent Electrical Systems (2ISEI) Laboratory High School of Technology (ESTM) University Moulay Ismail Meknes Morocco

ISBN: (数字)9798331532970

ISBN: (纸本)9798331532987

This study focuses on the exploration of five novel π-conjugated organic compounds for potential incorporation into organic light-emitting diodes (OLEDs). The molecular structure of these compounds follows a D-Di-D setup, with a carbazole acting as the D component, and the specific Di component varying based on the donor utilized. Through analysis at the DFT/B3LYP/6-31G (d, p) level, including the determination of HOMO–LUMO energy values, after optimizing the molecules' ground state, the frontier molecular orbital (FMO) was studied using Egap band gap. Utilizing the TD-DFT and ZINDO method the optical properties like absorption (λmax) and emission (λem) were explored. Based on SILVACO-TCAD software, the estimated current-voltage (I-V) characteristics of the two most efficient molecules were obtained. The indices of the studied molecules calculated from these interactions indicates that the molecules have sufficient properties which allow these molecules to be considered promising candidates for use in OLED devices as high-performing active materials.

关键词： Photonic band gap Absorption Stimulated emission Stationary state Organic light emitting diodes Market research Software Orbits Optical devices Organic compounds

来源：评论

学校读者我要写书评

暂无评论

Hybrid Blockchain and IoT Driven Framework for Supply Chain Optimization by Enhancing Transparency, Efficiency and Cost Reduction 3

Hybrid Blockchain and IoT Driven Framework for Supply Chain ...

引用

3rd International Conference on Electronics and Renewable systems, ICEARS 2025

作者： Pradeep Kumar, S. Basha, C.H. Hussaian Prasath, C. Arun Amirthayogam, G. Lathif, Syed Ismail Abdul Leelavathi, R. Rajalakshmi Engineering College Computer Science Design Department Thandalam Chennai602105 India Sr University Department of Electrical and Electronics Engineering Warangal506371 India Mahendra Engineering College Department of Ece Namakkal637503 India Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology School of Computing Department of Computer Science and Engineering Chennai600062 India Srm Institute of Science and Technology Department of Data Science and Business Systems Chennai603203 India Velalar College of Engineering and Technology Department of Biomedical Engineering Erode638012 India

ISBN: (纸本)9798331509675

Blockchain technology represents a modern supply chain management revolution which fixes operational challenges while providing enhanced transparency at reduced costs. The research develops HB-SCOF (Hybrid Blockchain-Enabled Supply Chain Optimization Framework) which combines blockchain self-verification databases with automatic smart contracts to maintain continuous supply chain observation and evaluate data credibility. The framework integrates tamper-proof transaction tracking mechanisms with unified supply chain network connectivity for securing data transmission as a trust delivery method through unbreakable shared databases. The retail-industry case study demonstrated a 25% decrease in expenses stemming from agency elimination alongside enhanced flow management and better stock distribution. The system demonstrates its effectiveness through assessments of key performance indicators including transaction speed and data accuracy together with business cost efficiency. This part looks at implementation methods through training stakeholders in addition to transforming organizational policies which produce successful outcomes. Blockchain technology labs verify its capability to enhance supply chain management efficiency while showing realistic deployment methods. © 2025 IEEE.

关键词： Smart contract

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：