检索结果-内蒙古大学图书馆

Attar: RRAM-based in-memory attention accelerator with software-hardware co-optimization

Science China(information Sciences) 2025年第3期68卷 371-387页

作者： Bing LI Ying QI Ying WANG Yinhe HAN Information Engineering College Capital Normal University Research Center for Intelligent Computing Systems Institute of Computing TechnologyChinese Academy of Sciences

The attention mechanism has become a pivotal component in artificial intelligence, significantly enhancing the performance of deep learning applications. However, its quadratic computational complexity and intricate computations lead to substantial inefficiencies when processing long sequences. To address these challenges, we introduce Attar, a resistive random access memory(RRAM)-based in-memory accelerator designed to optimize attention mechanisms through software-hardware co-optimization. Attar leverages efficient Top-k pruning and quantization strategies to exploit the sparsity and redundancy of attention matrices, and incorporates an RRAM-based in-memory softmax engine by harnessing the versatility of the RRAM crossbar. Comprehensive evaluations demonstrate that Attar achieves a performance improvement of up to 4.88× and energy saving of 55.38% over previous computing-in-memory(CIM)-based accelerators across various models and datasets while maintaining comparable accuracy. This work underscores the potential of in-memory computing to enhance the efficiency of attention-based models without compromising their effectiveness.

关键词： RRAM computing-in-memory attention pruning quantization

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence for Secured information systems in Smart Cities: Collaborative IoT computing with Deep Reinforcement Learning and Blockchain

arXiv

引用

arXiv 2024年

作者： Far, Amin Zakaie Far, Mohammad Zakaie Gharibzadeh, Sonia Naeini, Hajar Kazemi Amini, Leila Zangeneh, Shiva Rahimi, Morteza Asadi, Saeed Department of Mechanical Engineering Politecnico di Milano Milano Italy Department of Computer Engineering South Tehran Branch Islamic Azad University Tehran Iran Department of Civil Engineering University of Texas at Arlington ArlingtonTX United States Department of Information Systems and Business Analytics Florida International University MiamiFL United States Faculty of Engineering University of Malayer Malayer Iran School of Computing and Information Sciences Florida International University MiamiFL United States

The accelerated expansion of the Internet of Things (IoT) has raised critical challenges associated with privacy, security, and data integrity, specifically in infrastructures such as smart cities or smart manufacturing. Blockchain technology provides immutable, scalable, and decentralized solutions to address these challenges, and integrating deep reinforcement learning (DRL) into the IoT environment offers enhanced adaptability and decision-making. This paper investigates the integration of blockchain and DRL to optimize mobile transmission and secure data exchange in IoT-assisted smart cities. Through the clustering and categorization of IoT application systems, the combination of DRL and blockchain is shown to enhance the performance of IoT networks by maintaining privacy and security. Based on the review of papers published between 2015 and 2024, we have classified the presented approaches and offered practical taxonomies, which provide researchers with critical perspectives and highlight potential areas for future exploration and research. Our investigation shows how combining blockchain's decentralized framework with DRL can address privacy and security issues, improve mobile transmission efficiency, and guarantee robust, privacy-preserving IoT systems. Additionally, we explore blockchain integration for DRL and outline the notable applications of DRL technology. By addressing the challenges of machine learning and blockchain integration, this study proposes novel perspectives for researchers and serves as a foundational exploration from an interdisciplinary standpoint. Copyright © 2024, The Authors. All rights reserved.

关键词： Smart manufacturing

来源：评论

学校读者我要写书评

暂无评论

Towards Ethical AI Moderation in Multiplayer Games

引用

Proceedings of the ACM on Human-Computer Interaction 2024年第CHI PLAY期8卷 1-30页

作者： Sparrow, Lucy A. Galwey, Ren Jovic, Dahlia Hardwick, Taylor Butt, Mahli-Ann School of Computing and Information Systems The University of Melbourne Australia

AI is increasingly being used to moderate player behaviour in online multiplayer games, working to identify and respond to toxic and problematic conduct with greater efficiency and accuracy than existing automated systems. However, little work has explored the application of AI moderation in the gaming ecosystem, despite growing ethical concerns about AI applications in other domains. In this study, we conducted 2 expert workshops and interviewed 26 players and industry professionals on their understandings, perceptions and experiences with AI moderation in multiplayer games. Applying a metaphorical frame via template analysis, we outline four metaphors that capture participants’ views on the roles of AI and automation in moderation: the Unreliable Police Force, the Unscrupulous Governor, the Uncaring Judge, and the Untiring Assistant. We discuss these roles as exacerbating a top-down, punitive online justice system and identify ethical concerns around transparency, fairness and inclusion, privacy, and human-AI collaboration. To address these concerns, we put forward a set of ethical design considerations and alternative roles for AI moderation in multiplayer games. © 2024 Copyright is held by the owner/author(s).

关键词： Game design

来源：评论

学校读者我要写书评

暂无评论

Time Series Classification with Elasticity Using Augmented Path Signatures

引用

ACM Transactions on Knowledge Discovery from Data 2025年第3期19卷 1-32页

作者： Wang, Chenyang Luo, Ling Aickelin, Uwe The University of Melbourne School of Computing and Information Systems Parkville Australia

We often compare time-dependent data elastically such that some compression or dilation along the time dimension can be ignored, for example, spatial trajectories of vehicles moving at different speeds or accelerometer data for exercises completed at variable rhythms. Traditionally this is possible via an alignment-based elastic distance measure, such as dynamic time warping (DTW). We may also control the degree of allowable warping with warping constraints. However, these elastic distance measures are not easy to use in large-scale time series classification, as they need to be evaluated pairwise and often cannot be directly converted into feature sets that we may use with arbitrary classifiers or combine with other features. In this research, we focus on the study of path signatures, a transformation with time warping invariance property, and how we may augment a time series to make its signature space representation reflect common warping constraints. We demonstrate that the comparing signatures is analogous to comparing time series with elastic distances, and that augmented signature features can serve as warping invariant or insensitive features in time series classification. Finally, we construct multiple path signatures with constraining augmentations classifier (MultiPSCA), a general-purpose minimal tuning time series classifier using augmented signatures and show that it is able to beat existing best-performing elastic time series classification algorithms without per-dataset hyperparameter tuning. © 2025 Copyright held by the owner/author(s).

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

Optimizing Medical Image Analysis: A Performance Evaluation of YOLO-Based Segmentation Models

引用

International Journal of Advanced Computer Science and Applications 2025年第4期16卷 1167-1174页

作者： Alanazi, Haifa Department of Information Systems-Faculty of Computing and Information Technology Northern Border University Saudi Arabia

Instance segmentation is a critical component of medical image analysis, enabling tasks such as tissue and organ delineation, and disease detection. This paper provides a detailed comparative analysis of two fine-tuned one-stage object detection models, YOLOv11-seg and YOLOv9-seg, tailored for instance segmentation in medical imaging. Leveraging transfer learning, both models were initialized with pretrained weights and sub sequently fine-tuned on the NuInsSeg dataset, which comprises over 30,000 manually segmented nuclei across 665 image patches from various human and mouse organs. This approach facilitated faster convergence and improved generalization, particularly given the limited size and high complexity of the medical dataset. The models were evaluated against key performance metrics. The experimental results reveal that YOLOv11n-seg outperforms YOLOv9c-seg with a precision of 0.87, recall of 0.84, and mAP50 of 0.89, indicating superior segmentation quality and more accurate delineation of nuclei contours. This study highlights the robust performance and efficiency of YOLOv11n seg, demonstrating its superiority in medical image segmentation tasks, with notable advantages in both accuracy and real-time processing capabilities. © (2025), (Science and information Organization). All Rights Reserved.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Pr-Ge-Ne: Efficient Encoding of Pervasive Video Sensing Streams by Pruned Generative Networks

引用

ACM Transactions on Multimedia computing, Communications and Applications 2025年第2期21卷 1-22页

作者： Wu, Ji-Yan Gamlath, Kasun Misra, Archan School of Computing and Information Systems Singapore Management University Singapore Singapore

While video sensing, performed by resource-constrained pervasive devices, is a key enabler of many machine intelligence applications, the high energy and bandwidth overheads of streaming video transmission continue to present formidable deployment challenges. Motivated by the recent advancements in deep learning models, this article proposes the usage of a Generative Network-based technique for resource-efficient streaming video compression and transmission. However, we empirically show that while such generative network-based models offer superior compression gains compared to H.265, additional DNN optimization mechanisms are needed to substantially reduce their encoder complexity. Our proposed optimized system, dubbed Pr-Ge-Ne, adopts a carefully pruned encoder-decoder DNN, on the pervasive device, to efficiently encode a latent vector representation of intra-frame relative motion, and then uses a generator network at the decoder to reconstruct the frames by overlaying such motion information to "animate"an initial reference frame. By evaluating three representative streaming video datasets, we show that Pr-Ge-Ne achieves around - fold reduction in video transmission rates (with negligible impact on the accuracy of machine perception tasks) compared to H.265, while simultaneously reducing latency and energy overheads on a pervasive device by 90% and 15-50%, respectively. © 2025 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Encoding (symbols)

来源：评论

学校读者我要写书评

暂无评论

Pervasive Attentive Neural Network for Intelligent Image Classification Based on N-CDE’s

引用

Computers, Materials & Continua 2024年第4期79卷 1137-1156页

作者： Anas W.Abulfaraj Department of Information Systems Faculty of Computing and Information TechnologyKing Abdulaziz UniversityRabigh21911Saudi Arabia

The utilization of visual attention enhances the performance of image classification *** attentionbased models have demonstrated notable performance,but many of these models exhibit reduced accuracy when confronted with inter-class and intra-class similarities and ***-Controlled Differential Equations(N-CDE’s)and Neural Ordinary Differential Equations(NODE’s)are extensively utilized within this ***’s possesses the capacity to effectively illustrate both inter-class and intra-class similarities and differences with enhanced *** this end,an attentive neural network has been proposed to generate attention maps,which uses two different types of N-CDE’s,one for adopting hidden layers and the other to generate attention *** distinct attention techniques are implemented including time-wise attention,also referred to as bottom N-CDE’s;and element-wise attention,called topN-CDE’***,a trainingmethodology is proposed to guarantee that the training problem is sufficiently *** classification tasks including fine-grained visual classification andmulti-label classification,are utilized to evaluate the *** proposedmethodology is employed on five publicly available datasets,including CUB-200-2011,ImageNet-1K,PASCAL VOC 2007,PASCAL VOC 2012,and MS *** obtained visualizations have demonstrated that N-CDE’s are better appropriate for attention-based activities in comparison to conventional NODE’s.

关键词： Differential equations neural-controlled DE image classification attention maps N-CDE’s

来源：评论

学校读者我要写书评

暂无评论

Dawwen: An Arabic Mental Health Mobile App Based on Natural Language Processing

International Journal of Interactive Mobile Technologies

引用

International Journal of Interactive Mobile Technologies 2025年第4期19卷 108-131页

作者： Wali, Arwa Almagrabi, Hana El-Feky, Sarah Jokhdar, Maryah Department of Information Systems Faculty of Computing and Information Technology King Abdulaziz University Jeddah Saudi Arabia

People are increasingly concerned about their mental health wellness. Scientific studies suggest that online counselling for anxiety and depression is just as effective as in-person treatment. Additionally, journaling interventions have shown promise for individuals dealing with mental and psychological issues. In recent years, a growing number of mobile applications have been developed to improve people’s mental wellness and emotional commu-nication. However, many of these applications are not available in Arabic but are available only in English or the native languages of their users, while other applications have feature limitations. This study presents a prototype of an interactive mental health mobile application, called Dawwen, to assist Arab individuals in freely expressing their feelings through online journaling, receiving recommendations for practices and activities, and locating nearby therapy centers based on their geolocation. Dawwen is augmented with an easy-to-use interface, a natural language processing (NLP) technique for analyzing the user input, and integrated features. The system is implemented using Android Studio and various cloud-based tools, making it adaptable for the integration of more advanced artificial intelligence (AI) technologies in the future. The System Usability Scale (SUS) scored an average of 88.25%, indicat-ing high user satisfaction with the app’s functionality and ease of use. The study highlights Dawwen’s effectiveness in improving mental health support for Arabic-speaking users, filling a critical gap in existing digital mental health resources. © 2025 by the authors of this article. Published under CC-BY.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Clustering Analysis of Physicians’ Performance Evaluation: A Comparison of Feature Selection Strategies to Support Medical Decision-Making

引用

International Journal of Advanced Computer Science and Applications 2025年第4期16卷 700-707页

作者： Ghazzawi, Amani Mustafa Almagrabi, Alaa Omran Namankani, Hanaa Mohammed Department of Information Systems Faculty of Computing and Information Technology King Abdulaziz University Jeddah. Saudi Arabia Department of Management Information Systems Faculty of Business Administration Taif University Taif Saudi Arabia Department of Information Systems Faculty of Computing and Information Technology King Abdulaziz University Jeddah Saudi Arabia

Evaluating physicians’ performance is one of the fundamental pillars of improving the quality of healthcare in medical institutions, as it contributes to measuring their ability to provide appropriate treatment, interact effectively with patients, and work within healthcare teams. This study aims to explore the impact of attribute selection on the accuracy of physician clustering using the K-Means algorithm, to improve physician performance assessment. Three datasets containing professional, medical, and administrative attributes were analyzed, such as age, nationality, job title, years of experience, number of operations, and evaluations from various entities. The optimal number of clusters was determined using the Elbow and Silhouette Score methods. The results showed that the original feature set and Lasso features performed best at k = 3, with a clear distinction between clusters. The "three-star" cluster performed well at k = 2 but lost some fine details. It was also shown that attribute selection directly affects the number and accuracy of clusters resulting from clustering, allowing for a clearer classification of physician categories. The study recommends using either original features or Lasso features to achieve more effective clustering, which supports improved recruitment, training, and management decision-making processes in healthcare organizations. © (2025), (Science and information Organization). All rights reserved.

关键词： Risk perception

来源：评论

学校读者我要写书评

暂无评论

Enhancing Ulcerative Colitis Diagnosis:A Multi-Level Classification Approach with Deep Learning

引用

Computer Modeling in Engineering & Sciences 2024年第7期140卷 1129-1142页

作者： Hasan J.Alyamani Department of Information Systems Faculty of Computing and Information Technology in RabighKing Abdulaziz UniversityJeddah21589Saudi Arabia

The evaluation of disease severity through endoscopy is pivotal in managing patients with ulcerative colitis,a condition with significant clinical ***,endoscopic assessment is susceptible to inherent variations,both within and between observers,compromising the reliability of individual *** study addresses this challenge by harnessing deep learning to develop a robust model capable of discerning discrete levels of endoscopic disease *** initiate this endeavor,a multi-faceted approach is embarked *** dataset is meticulously preprocessed,enhancing the quality and discriminative features of the images through contrast limited adaptive histogram equalization(CLAHE).A diverse array of data augmentation techniques,encompassing various geometric transformations,is leveraged to fortify the dataset’s diversity and facilitate effective feature extraction.A fundamental aspect of the approach involves the strategic incorporation of transfer learning principles,harnessing a modified ResNet-50 *** augmentation,informed by domain expertise,contributed significantly to enhancing the model’s classification *** outcome of this research endeavor yielded a highly promising model,demonstrating an accuracy rate of 86.85%,coupled with a recall rate of 82.11%and a precision rate of 89.23%.

关键词： Ulcerative colitis deep learning CLAHE transfer learning geometric transformations

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：