检索结果-内蒙古大学图书馆

International Conference on Computing and Networking Technology (ICCNT)

作者： N Raghu Shreya Chakraborty Aakriti Chauhan Department of Electrical and Electronics Engineering School of Engineering and Technology JAIN Bangalore Karnataka India Department of optometry ARKA JAIN University Jamshedpur Jharkhand India Department of Agri-Business Management Vivekananda Global University Jaipur India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

Herbal Language Processing (NLP) and Mining algorithms are frequently used to research big collections of patient electronic fitness information (EHR) to acquire insights that could be beneficial to patient care. Mining algorithms can assist to automate the technique of reading huge amounts of patient facts and may provide useful facts from affected person facts that would in any other case be difficult to method. NLP is getting used to extract records from unstructured textual content in patient statistics, that may useful resource in determining sicknesses, symptoms, medicines, and chance elements. by using the usage of NLP and system studying algorithms, all the facts from affected person records may be efficiently and as it should be processed, resulting in study’s findings which can assist improve healthcare. NLP and mining algorithms for affected person facts can assist to streamline disorder diagnosis, provide extra accurate records approximately clinical selections, and ultimately improve patient results.

关键词： Accuracy Databases Medical services Companies approximation algorithms Natural language processing Data mining Medical diagnostic imaging

来源：评论

学校读者我要写书评

暂无评论

Near-field Three-dimensional Imaging of Large MIMO Arrays Based on Spatial Phase Compensation

Near-field Three-dimensional Imaging of Large MIMO Arrays Ba...

引用

International Conference on Electronics and Information Technology (EIT)

作者： Zhenbiao Liao Hanfei Zhou Wenlong He College of Electronic and Information Engineering Shenzhen University Shenzhen China

ISBN: (数字)9798350369151

ISBN: (纸本)9798350369168

The growing interest in active millimeter-wave and microwave imaging systems is driven by their applications in security, healthcare, and industrial fields. However, imaging systems using single antenna arrays suffer from low efficiency, and Multi-Input Multi-Output (MIMO) radar imaging systems often lack far-field conditions at short distances. In this paper, we propose a novel MIMO near-field imaging algorithm specifically designed for large millimeter-wave radar systems. By employing a linear approximation to compensate for the spatial quadratic phase at a single reference point, the algorithm enables the application of FFT-based imaging techniques within a defined range. The paper details the imaging range applicable to a single reference point, as well as the calculation method and imaging steps for multiple reference points in global imaging. Experimental results demonstrate that the proposed algorithm effectively focuses target energy, enabling real-time 3D near-field imaging for real aperture arrays.

关键词： Phased arrays Three-dimensional displays MIMO radar Imaging Linear approximation Radar imaging Millimeter wave radar approximation algorithms Real-time systems Security

来源：评论

学校读者我要写书评

暂无评论

Computational Limits of Low-Rank Adaptation (LoRA) for Transformer-Based Models

arXiv

引用

arXiv 2024年

作者： Hu, Jerry Yao-Chieh Su, Maojiang Kuo, En-Jui Song, Zhao Liu, Han Department of Computer Science Northwestern University EvanstonIL60208 United States Department of Information and Computing Science USTC Anhui Hefei230026 China Physics Division National Center for Theoretical Sciences Taipei106319 Taiwan Adobe Research SeattleWA98103 United States Department of Statistics and Data Science Northwestern University EvanstonIL60208 United States

We study the computational limits of Low-Rank Adaptation (LoRA) update for finetuning transformer-based models using fine-grained complexity theory. Our key observation is that the existence of low-rank decompositions within the gradient computation of LoRA adaptation leads to possible algorithmic speedup. This allows us to (i) identify a phase transition behavior and (ii) prove the existence of nearly linear algorithms by controlling the LoRA update computation term by term, assuming the Strong Exponential Time Hypothesis (SETH). For the former, we identify a sharp transition in the efficiency of all possible rank-r LoRA update algorithms for transformers, based on specific norms resulting from the multiplications of the input sequence X, pretrained weights W*, and adapter matrices αBA/r. Specifically, we derive a shared upper bound threshold for such norms, and show that efficient (sub-quadratic) approximation algorithms of LoRA exist only below this threshold. For the latter, we prove the existence of nearly linear approximation algorithms for LoRA adaptation by utilizing the hierarchical low-rank structures of LoRA gradients and approximating the gradients with a series of chained low-rank approximations. To showcase our theory, we consider two practical scenarios: partial (e.g., only WV and WQ) and full adaptations (e.g., WQ, WV, and WK) of weights in attention heads. © 2024, CC BY-NC-SA.

关键词： approximation algorithms

来源：评论

学校读者我要写书评

暂无评论

Merging and cooperative caching based HDFS small files access performance optimization

Merging and cooperative caching based HDFS small files acces...

引用

Communication Technology and Information Technology (ICCTIT), International Conference on

作者： Weijing Xu Yitao Qiu Congfeng Jiang Junming Liu Xiaoyu Wang Shijie Chen School of Computer Science and Technology Hangzhou Dianzi University Hangzhou China

ISBN: (数字)9798331528973

ISBN: (纸本)9798331528980

In this paper, we propose a performance optimization method for accessing small files in HDFS based on merging an d co-caching strategies to address the memory overhead problem caused by massive small files in Hadoop Distributed File System (HDFS). First, the small files stored in HDFS are merged into multiple large files based on the descending first adaptation algorithm, thus saving the memory space occupied by the management node (NameNode). Second, in order to improve the read speed of s mall files, this paper proposes a two-level collaborative caching strategy, i.e., the server-side caching adopts the improved GDSF (G reedy Dual Size Frequency) strategy for small files, and the client -side adopts the ARC (Adaptive Replacement Cache) caching strategy for the small file indexes. The Experiments show that compared with genetic algorithm and particle swarm algorithm, the number of large files generated by merging small files using descending first adaptation algorithm is reduced by 26 and 60 on average. Compared with the original GDSF strategy, the co-caching strategy proposed in this paper improves the server-side cache hit rate by 8.2% on average. Compared with the FIFO and LRU cache re placement strategies, the client-side cache hit ratio is improved by 2.7% and 8.9%, respectively.

关键词： Heuristic algorithms Merging Memory management Collaboration Metadata approximation algorithms Servers Particle swarm optimization Information technology Optimization

来源：评论

学校读者我要写书评

暂无评论

Querying Shortest Path on Large Time-Dependent Road Networks with Shortcuts

Querying Shortest Path on Large Time-Dependent Road Networks...

引用

International Conference on Data Engineering

作者： Zengyang Gong Yuxiang Zeng Lei Chen The Hong Kong University of Science and Technology Hong Kong SAR China School of Computer Science and Engineering Beihang University DSA Thrust The Hong Kong University of Science and Technology (Guangzhou) Guangzhou China

ISBN: (数字)9798350317152

ISBN: (纸本)9798350317169

Querying the shortest path between two locations is a fundamental task in many applications, and has been extensively studied for static road networks. However, in reality, the travel costs of road segments evolve over time, and hence the road network can be modeled as a time-dependent graph. In this paper, we study the shortest path query over large-scale time-dependent road networks. We first present a tree decomposition method to model the time-dependent road network as a tree structure that preserves travel costs. To further improve query efficiency, a set of shortcuts is selected and built on the constructed tree structure. Specifically, we formally define a shortcut selection problem over the tree decomposition of the time-dependent road network. This problem, which is proven to be NP-hard, aims to select and build the most effective shortcut set. We first devise a dynamic programming method with exact results to solve the selection problem. To obtain the optimal shortcut set quickly, we design an approximation algorithm that guarantees a 0.5-approximation ratio. Based on the novel tree structure, we devise a shortcut-based algorithm to answer the shortest path query over time-dependent road networks. Finally, we conduct extensive performance studies using large-scale real-world road networks. The results demonstrate that our method can achieve better efficiency and scalability than the state-of-the-art method.

关键词： Costs Roads Heuristic algorithms Scalability Query processing approximation algorithms Dynamic programming

来源：评论

学校读者我要写书评

暂无评论

DCP++: A New Registration Method for Brain MRI

DCP++: A New Registration Method for Brain MRI

引用

Medical Artificial Intelligence (MedAI), IEEE International Conference on

作者： Yucheng Liao Bohan Lei Yueting Zhuang Xiaoyin Xu Min Zhang Chu Kochen Honors College Zhejiang University Hangzhou China College of Computer Science & Technology Zhejiang University Hangzhou China College of Biomedical Engineering & Instrument Science Zhejiang University Hangzhou China

ISBN: (数字)9798350377613

ISBN: (纸本)9798350377620

In geometric computer vision, 3D registration is a critical task for various applications. 3D medical image registration, as one important application of 3D registration, holds substantial clinical potentials for supporting diagnosis, surgeries, and treatments through the registration of images captured at different times or using different modalities. Recently, the rapid advancement and widespread adoption of deep learning have spurred interest in applying this technology to the registration of point cloud data. In this paper, an deep-learning-based 3D brain MRI medical data registration algorithm (DCP++) is constructed. Through employing reasonable loss function, DCP++ achieves efficient and precise registration of 3D brain data. To evaluate the performance of DCP++, we apply the algorithm to the point cloud registration task of 3D human brain MRI images from the ADNI public dataset. Benchmark comparisons, ablation experiment, and visualization analyses are conducted. Experiments demonstrate that compared with state-of-the-art algorithms, DCP++ reduces the registration error by approximately 90%. Tests on real data confirm the accuracy and robustness of the DCP++ algorithm, highlighting its potential in medical practice.

关键词： Point cloud compression Image registration Three-dimensional displays Accuracy Magnetic resonance imaging Surgery Benchmark testing approximation algorithms Robustness Medical diagnostic imaging

来源：评论

学校读者我要写书评

暂无评论

Data-Driven Event-Triggered Resilient Control for Vehicular Platooning Systems Under False Data Injection Attacks

Data-Driven Event-Triggered Resilient Control for Vehicular ...

引用

International Conference on Electronic Engineering and Informatics (EEI)

作者： Shanshan Sun Kai Zhou Yueming Wang Yuanxin Li The College of Science Liaoning University of Technology Jinzhou China

ISBN: (数字)9798350353594

ISBN: (纸本)9798350353600

The article explores the event-triggered resilient model-free adaptive platooning security control (MFAPC) problem for vehicular platooning systems (VPSs) with asymmetric constraints and false data injection attacks (FDI). Firstly, via the system transformation, an unconstrained function expression of the controlled system is developed to overcome the challenge caused by asymmetric constraints of position and velocity. Secondly, based on transformed functions, an equivalent linear model can be acquired from the nonlinear VPSs by virtue of the dynamic linearization approach. Thirdly, taking the excellent performance of the function approximation into account, the radial basis neural network is introduced to approximate the false data, which significantly reduces the negative effects of attacks. Moreover, an event-triggered mechanism is introduced to save network resources effectively. For the VPSs subjected to FDI attacks, a new resilient ET-MFAPSC scheme is proposed for safe vehicle driving. In conclusion, the effectiveness of the designed algorithm can be demonstrated through a simulation example.

关键词： Adaptation models Adaptive systems Neural networks Vehicle driving Control systems approximation algorithms Data models Nonlinear dynamical systems Function approximation Vehicle dynamics

来源：评论

学校读者我要写书评

暂无评论

Algorithm for Load Balancing of a Data Processing Center Based on a Nonlinear Forecast Model

Algorithm for Load Balancing of a Data Processing Center Bas...

引用

International Russian Automation Conference (RusAutoCon)

作者： Valery P. Mochalov Natal'ya Yu. Bratchenko Dar'ya V. Gosteva Chair of infocommunications North-Caucasus Federal University Stavropol Russia

ISBN: (数字)9798350349818

ISBN: (纸本)9798350349825

The purpose of the article is to increase the efficiency of the cloud data center load distribution and balancing system by developing and applying a mechanism for predicting network traffic conditions characterized by fractal self-similarity. When developing the predictive model, methods of nonlinear dynamics were used, taking into account the statistical self-similarity of the load and providing a solution to the problem of predicting the moments of its expected bursts. The randomness of network traffic was checked by calculating the spectrum of Lyapunov exponents. The mathematical model and the dynamic algorithm of the predictive model of the state of a nonlinear system are presented in the form of a system of discrete mappings of previous and subsequent values of a time series and a regression approximating polynomial connecting them. The presented algorithm differs from the existing ones by taking into account the features of fractal self-similarity of the input load, which negatively affects quality indicators, using a predictive model developed on the basis of nonlinear dynamics methods, as well as the possibility of choosing rational balancing parameters according to the criterion of uniform loading of server resources. The article shows that the dynamic load balancing algorithm, based on nonlinear approaches and predictive models, allows solving load distribution problems between servers of data center clusters more efficiently than traditional methods. The conclusion about the effectiveness of the developed algorithm is substantiated.

关键词： Data centers Heuristic algorithms Time series analysis Telecommunication traffic Predictive models Prediction algorithms approximation algorithms Fractals Nonlinear dynamical systems Servers

来源：评论

学校读者我要写书评

暂无评论

Runtime Configurable Approximate Computing System for Simulated Annealing Algorithm

Runtime Configurable Approximate Computing System for Simula...

引用

China Semiconductor Technology International Conference (CSTIC)

作者： Jian Shi Weikang Qian University of Michigan-SJTU Joint Institute China MoE Key Lab of AI Shanghai Jiao Tong University China

ISBN: (数字)9798350362190

ISBN: (纸本)9798350362206

In this paper, we propose a runtime configurable approximate computing (RCAC) system for simulated annealing (SA) algorithm. The RCAC system integrates a configurable approximate hardware to calculate the cost function and a lightweight quality estimator, which provides valuable insights into the SA algorithm's effectiveness by estimating the acceptance rate of each iteration. By utilizing the acceptance rate, the system dynamically configures the operational mode of the approximate hardware, making a good trade-off between solution quality and computational efficiency. Experimental results demonstrate a 7.6% runtime improvement compared to traditional RCAC methods on SA algorithms with a quality loss of less than 2% in final solutions.

关键词： Measurement Integrated circuits Runtime Heuristic algorithms Approximate computing Simulated annealing approximation algorithms

来源：评论

学校读者我要写书评

暂无评论

Line Simplification for Efficient Approximate Join Queries On Big Geospatial Data

Line Simplification for Efficient Approximate Join Queries O...

引用

Intelligent Data Science Technologies and Applications (IDSTA), International Conference on

作者： Fatima Ahmed Alhammadi Haya Almadhloum Alsuwaidi Shooq Abdelrahman Alzarooni Isam Mashhour Al Jawarneh Department of Computer Science University of Sharjah P.O.Box. Sharjah United Arab Emirates

ISBN: (数字)9798350354751

ISBN: (纸本)9798350354768

Line simplification algorithms are often used to render high-resolution geographic features at appropriate resolutions when applied to polygons. They are generalization techniques in which selective vertices are removed from a line feature to eliminate details whilst preserving the line’s basic shape. In this paper, two different line simplification algorithms (Douglas-Peucker and Visvalingam-Whyatt) are used in conjunction with spatial join of geo-referenced mobility and air quality data, to reduce the size of the polygon files. A filter-and-refinement dimensionality reduction-based approach is then used to join the data. This framework allows for an optimized spatial join on an integrated schema through a state-of-the-art filter-and-refine based approach. The reduced files can then be used in geospatial related data science tasks such as DBSCAN, clustering, and regression at lower computational costs. Our experimental results show that incorporating a reduction approach such as line simplification before performing the spatial join can significantly reduce the computational cost and improve the performance with the number of vertices reduced by 94% after simplification and accuracy, MAPE is minimized with a low score of 0.048 and 0.049 for DP-Map shaper and VW-Map shaper respectively.

关键词： Accuracy Shape Data science Filtering algorithms approximation algorithms Air quality Geospatial analysis Computational efficiency Spatial resolution

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：