检索结果-内蒙古大学图书馆

Emerging Trends in Electrical, Control, and Telecommunication Engineering (ETECTE), International Conference on

作者： Abeera Malik Faisal Bukhari Hafiz Muhammad Talha Department of Information Technology Faculty of Computing and Information Technology University of the Punjab PUCIT Lahore Pakistan Department of Data Science Faculty of Computing and Information Technology University of the Punjab Lahore Pakistan

ISBN: (数字)9798331504861

ISBN: (纸本)9798331504878

Several industries have embraced blockchain technology due to the change it has brought, especially in the way records are kept. However, one major problem inherent in the concept of blockchain from the participants perspective is the presence of unpredictable rewards in proof-of-work and proof- of-stake models. That is why this research aims at developing a new strategy to solve this problem by incorporating artificial intelligence (AI) as well as machine learning (ML) algorithms into blockchain compensation prediction. This project seeks to analyze how artificial intelligence (AI) and machine learning (ML) methods may be incorporated into blockchain technology to solve the problem of the unpredictability of rewards. Using the EtherScan API, we have obtained 1000 timestamps and block reward values, and for prediction of rewards for certain timestamps, we use the KNN algorithm, linear regression algorithm, and random forest regressor algorithm. Thus, a What We KNN attained an accuracy of performance specifies, and a random forest regressor attained an impression accuracy range of 74% particular timestamps. These outcomes show the benefits of advanced ML to the blockchain ecosystem as a tool that helps stakeholders make informed decisions concerning activities like mining, staking, or investing in blockchains with potentially high rewards.

关键词： Technological innovation Machine learning algorithms Accuracy Biological system modeling Nearest neighbor methods Prediction algorithms Proof of Work Blockchains Stakeholders Random forests

来源：评论

学校读者我要写书评

暂无评论

Prediction of Sprint Delivery Capability in Agile Software Development Using Machine Learning

Prediction of Sprint Delivery Capability in Agile Software D...

引用

International Research Conference on Smart computing and Systems Engineering (SCSE)

作者： Narayanasamy Lorshan P. M. A. K. Wijerathna B. T. G. S. Kumara Department of Computing and Information Systems Faculty of Computing Sabaragamuwa University of Sri Lanka Belihuloya Sri Lanka Department of Software Engineering Faculty of Computing Sabaragamuwa University of Sri Lanka Belihuloya Sri Lanka Department of Data Science Faculty of Computing Sabaragamuwa University of Sri Lanka Belihuloya Sri Lanka

ISBN: (数字)9798331523268

ISBN: (纸本)9798331523275

Scrum is one of the most prominent agile development frameworks as it delivers the project in many sprints in an incremental manner. Sprint delivery capability is a crucial concept in the scrum projects as it assesses whether sprint can capable of delivering its issues on its length. It is difficult to consider forecasting sprint delivery capability in large scrum projects as multiple sprints are processed simultaneously and agile dynamics. Machine Learning (ML) models were applied in the existing studies to forecast sprint delivery but, had limitations in finding non-linear and hidden patterns. To address these problems, this research exploited the untapped ensemble learning approach to bridge the gap to improve the prediction of sprint delivery capability in different project progression levels. The study gathered the data from one of the large open-source scrum projects, JIRA which included 1873 iterations and 10853 issues in table format spanning at 30%, 50%, and 80% project progression levels. data preprocessing enabled to refinement of the data for analysis. The author enhanced the feature aggregation by using both statistical and automatic feature learning to derive the new features from the issue table and combine them with the iteration table. The key features affecting sprint delivery capability were identified using hybrid feature selection approach. The study developed the untapped ensemble models by choosing the different combinations of ML, Deep Learning (DL), and ensemble algorithms using appropriate ensemble methods and combination rules. The author examined the models and concluded that Categorical Boost (CatBoost) gained the highest accuracy of 96.8% at the 80% project progression level. This prediction assists in improving sprint planning, delivering the software on time, and getting more insights about sprint performance.

关键词： Representation learning Software algorithms Predictive models Feature extraction Prediction algorithms Software Planning Ensemble learning Modeling Forecasting

来源：评论

学校读者我要写书评

暂无评论

Machine Learning and OpenCV: A Powerful Combination for Driver Drowsiness Detection

Machine Learning and OpenCV: A Powerful Combination for Driv...

引用

2024 International Conference on Communication, computing and Internet of Things, IC3IoT 2024

作者： Siva Kumar, A. Gopirajan, P.V. Ramasamy, Jayaraj Jackson, Beulah School Of Computing Srm Institute Of Science And Technology Department Of Data Science And Business Systems Kattankulathur Chennai603203 India School Of Computing Srm Institute Of Science And Technology Department Of Computational Intelligence Kattankulathur Campus Chennai603203 India Botho University Faculty Of Engineering And Technology Department of IT Botswana Vel Tech Rangarajan Dr. Sagunthala R & D Institute Of Science And Technology Department Of Electronics And Communication Engineering Chennai India

ISBN: (纸本)9798350352689

Driver fatigue is a significant contributor to road accidents worldwide, and there is a need for efficient driver drowsiness detection systems to prevent such accidents. Computer vision techniques like OpenCV have recently been utilized to develop such strategies. This proposed work investigates the use of machine learning models and OpenCV Python package for driver drowsiness detection. The proposed system uses a camera to continuously capture the driver's face and applies image processing techniques to identify drowsiness. The system detects blinks by monitoring the driver's eyes and measuring the frequency and duration of eyelid closures. The real-time experiments conducted with the proposed method showed high accuracy in detecting drowsiness, and the system can be implemented in various settings, including cars and heavy vehicles. Ultimately, the system has some potential to reduce accidents caused by driver fatigue, thereby enhancing road safety. © 2024 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Topology Decoupled All-reduce Algorithm

Topology Decoupled All-reduce Algorithm

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Ruixing Zong Jiapeng Zhang Zhuo Tang Anwitaman Datta College of Computer Science and Electronic Engineering Hunan University China College of Computing and Data Science Nanyang Technological University Singapore Faculty of Computing Engineering and Media De Montfort University Leicester United Kingdom

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

With the advancement of deep learning, network communication has become the most critical factor in model training. Especially, the all-reduce operation can comprise over 70% of the cumulative training duration as a pivotal component within data parallelism. However, existing all-reduce algorithms often perform poorly in complex network topologies, and there is currently no universal and straightforward all-reduce algorithm that can effectively adapt to diverse topological *** this work, we propose a topology decoupled all-reduce algorithm. We decouple the network into multiple tree substructures, select the trees with the smallest heights, and then split the data to perform aggregate communication within these selected structures. This approach significantly reduces the number of communications and enhances efficiency. Experimental results show that our topology decoupled all-reduce algorithm reduces communication time compared to NCCL’s by 42.9% and enhances end-to-end training efficiency by 11.7%.

关键词： Training Deep learning Adaptation models Signal processing algorithms Complex networks Signal processing Parallel processing data aggregation Topology Speech processing

来源：评论

学校读者我要写书评

暂无评论

Deep Fourier-embedded Network for RGB and Thermal Salient Object Detection

arXiv

引用

arXiv 2024年

作者： Lyu, Pengfei Yeung, Pak-Hei Yu, Xiaosheng Wu, Chengdong Rajapakse, Jagath C. Faculty of Robot Science and Engineering Northeastern University Shenyang110169 China College of Computing and Data Science Nanyang Technological University 639798 Singapore

The rapid development of deep learning has significantly improved salient object detection (SOD) combining both RGB and thermal (RGB-T) images. However, existing deep learning-based RGB-T SOD models suffer from two major limitations. First, Transformer-based models with quadratic complexity are computationally expensive and memory-intensive, limiting their application in high-resolution bi-modal feature fusion. Second, even when these models converge to an optimal solution, there remains a frequency gap between the prediction and ground-truth. To overcome these limitations, we propose a purely Fourier transform-based model, namely Deep Fourier-Embedded Network (DFENet), for accurate RGB-T SOD. To address the computational complexity when dealing with high-resolution images, we leverage the efficiency of fast Fourier transform with linear complexity to design three key components: (1) the Modal-coordinated Perception Attention, which fuses RGB and thermal modalities with enhanced multi-dimensional representation;(2) the Frequency-decomposed Edge-aware Block, which clarifies object edges by deeply decomposing and enhancing frequency components of low-level features;and (3) the Fourier Residual Channel Attention Block, which prioritizes high-frequency information while aligning channel-wise global relationships. To mitigate the frequency gap, we propose Co-focus Frequency Loss, which dynamically weights hard frequencies during edge frequency reconstruction by cross-referencing bi-modal edge information in the Fourier domain. Extensive experiments on four RGB-T SOD benchmark datasets demonstrate that DFENet outperforms fifteen existing state-of-the-art RGB-T SOD models. Comprehensive ablation studies further validate the value and effectiveness of our newly proposed components. The code is available at https://***/JoshuaLPF/DFENet. Copyright © 2024, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Sparse Attention Decomposition Applied to Circuit Tracing

arXiv

引用

arXiv 2024年

作者： Franco, Gabriel Crovella, Mark Department of Computer Science Boston University United States Department of Computer Science Faculty of Computing & Data Sciences Boston University United States

Many papers have shown that attention heads work in conjunction with each other to perform complex tasks. It's frequently assumed that communication between attention heads is via the addition of specific features to token residuals. In this work we seek to isolate and identify the features used to effect communication and coordination among attention heads in GPT-2 small. Our key leverage on the problem is to show that these features are very often sparsely coded in the singular vectors of attention head matrices. We characterize the dimensionality and occurrence of these signals across the attention heads in GPT-2 small when used for the Indirect Object Identification (IOI) task. The sparse encoding of signals, as provided by attention head singular vectors, allows for efficient separation of signals from the residual background and straightforward identification of communication paths between attention heads. We explore the effectiveness of this approach by tracing portions of the circuits used in the IOI task. Our traces reveal considerable detail not present in previous studies, shedding light on the nature of redundant paths present in GPT-2. And our traces go beyond previous work by identifying features used to communicate between attention heads when performing IOI. © 2024, CC BY.

关键词： Signal encoding

来源：评论

学校读者我要写书评

暂无评论

The Frontier of data Erasure: Machine Unlearning for Large Language Models

arXiv

引用

arXiv 2024年

作者： Qu, Youyang Ding, Ming Sun, Nan Kanchana Zhu, Tianqing Niyato, Dusit Data61 Commonwealth Scientific and Industrial Research Organization Australia School of Systems & Computing University of New South Wales Australia School of Computer Science University of Sydney Australia Faculty of Data Science City University of Macau China College of Computing & Data Science Nanyang Technological University Singapore

Large Language Models (LLMs) are foundational to AI advancements, facilitating applications like predictive text generation. Nonetheless, they pose risks by potentially memorizing and disseminating sensitive, biased, or copyrighted information from their vast datasets. Machine unlearning emerges as a cutting-edge solution to mitigate these concerns, offering techniques for LLMs to selectively discard certain data. This paper reviews the latest in machine unlearning for LLMs, introducing methods for the targeted forgetting of information to address privacy, ethical, and legal challenges without necessitating full model retraining. It divides existing research into unlearning from unstructured/textual data and structured/classification data, showcasing the effectiveness of these approaches in removing specific data while maintaining model efficacy. Highlighting the practicality of machine unlearning, this analysis also points out the hurdles in preserving model integrity, avoiding excessive or insufficient data removal, and ensuring consistent outputs, underlining the role of machine unlearning in advancing responsible, ethical AI. © 2024, CC BY.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Analyzing the Convergence of Federated Learning with Biased Client Participation 9th

Analyzing the Convergence of Federated Learning with Bias...

引用

19th International Conference on Advanced data Mining and Applications, ADMA 2023

作者： Tan, Lei Hu, Miao Zhou, Yipeng Wu, Di School of Computer Science and Engineering Sun Yat-sen University Guangzhou510006 China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou510006 China School of Computing Faculty of Science and Engineering Macquarie University SydneyNSW2109 Australia

ISBN: (纸本)9783031466632

Federated Learning (FL) is a promising decentralized machine learning framework that enables a massive number of clients (e.g., smartphones) to collaboratively train a global model over the Internet without sacrificing their privacy. Though FL’s efficacy in non-convex problems is proven, its convergence amidst biased client participation lacks theoretical study. In this paper, we analyze the convergence of FedAvg on non-convex problems, which is the most renowned FL algorithm. We assume even data distribution but non-IID among clients, and elucidate the convergence rate of FedAvg in situations characterized by biased client participation. Our analysis reveals that biased client participation can significantly reduce the precision of the FL model. We validate this through trace-driven experiments, demonstrating that unbiased client participation results in 11% to 50% higher test accuracy compared to extremely biased client participation. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Advanced Arabic Alphabet Sign Language Recognition Using Transfer Learning and Transformer Models

arXiv

引用

arXiv 2024年

作者： Balat, Mazen Awaad, Rewaa Adel, Hend Zaky, Ahmed B. Aly, Salah A. CS & IT Dept. Egypt-Japanese University of Science & Technology Alexandria Egypt Faculty of Computing & Data Science Badya University Giza Egypt

This paper presents an Arabic Alphabet Sign Language recognition approach, using deep learning methods in conjunction with transfer learning and transformer-based models. We study the performance of the different variants on two publicly available datasets, namely ArSL2018 and AASL. This task will make full use of state-of-the-art CNN architectures like ResNet50, MobileNetV2, and EfficientNetB7, and the latest transformer models such as Google ViT and Microsoft Swin Transformer. These pre-trained models have been fine-tuned on the above datasets in an attempt to capture some unique features of Arabic sign language motions. Experimental results present evidence that the suggested methodology can receive a high recognition accuracy, by up to 99.6% and 99.43% on ArSL2018 and AASL, respectively. That is far beyond the previously reported state-of-the-art approaches. This performance opens up even more avenues for communication that may be more accessible to Arabic-speaking deaf and hard-of-hearing, and thus encourages an inclusive society. © 2024, CC0.

关键词： Transfer learning

来源：评论

学校读者我要写书评

暂无评论

Efficient Fourier Filtering Network with Contrastive Learning for UAV-based Unaligned Bi-modal Salient Object Detection

arXiv

引用

arXiv 2024年

作者： Lyu, Pengfei Yeung, Pak-Hei Cheng, Xiufei Yu, Xiaosheng Wu, Chengdong Rajapakse, Jagath C. The Faculty of Robot Science and Engineering Northeastern University Shenyang110169 China The College of Computing and Data Science Nanyang Technological University 639798 Singapore

Unmanned aerial vehicle (UAV)-based bi-modal salient object detection (BSOD) aims to segment salient objects in a scene utilizing complementary cues in unaligned RGB and thermal image pairs. However, the high computational expense of existing UAV-based BSOD models limits their applicability to real-world UAV devices. To address this problem, we propose an efficient Fourier filter network with contrastive learning that achieves both real-time and accurate performance. Specifically, we first design a semantic contrastive alignment loss to align the two modalities at the semantic level, which facilitates mutual refinement in a parameter-free way. Second, inspired by the fast Fourier transform that obtains global relevance in linear complexity, we propose synchronized alignment fusion, which aligns and fuses bi-modal features in the channel and spatial dimensions by a hierarchical filtering mechanism. Our proposed model, AlignSal, reduces the number of parameters by 70.0%, decreases the floating point operations by 49.4%, and increases the inference speed by 152.5% compared to the cutting-edge BSOD model (i.e., MROS). Extensive experiments on the UAV RGB-T 2400 and three weakly aligned datasets demonstrate that AlignSal achieves both real-time inference speed and better performance and generalizability compared to sixteen state-of-the-art BSOD models across most evaluation metrics. In addition, our ablation studies further verify AlignSal's potential in boosting the performance of existing aligned BSOD models on UAV-based unaligned data. The code is available at: https://***/JoshuaLPF/AlignSal. Copyright © 2024, The Authors. All rights reserved.

关键词： Unmanned aerial vehicles (UAV)

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：