检索结果-内蒙古大学图书馆

IEEE International Conference on Cloud Computing, CLOUD

作者： Seyed Morteza Nabavinejad Sherief Reda Tian Guo Computer Science Department Worcester Polytechnic Institute Worcester MA School of Engineering Brown University Providence RI

ISBN: (数字)9798350368536

ISBN: (纸本)9798350368543

With the increase in computing power of cutting-edge hardware platforms, it is a common practice to run multiple jobs on a single machine for improved resource utilization and throughput. However, this leads to inevitable resource contention among co-located jobs, impacting their performance. The resource contention can worsen due to fluctuations in resource utilization of jobs caused by variations in their input workload. To tackle the co-location contention for DNN inference jobs, we propose MediatorDNN, which considers contention and resource utilization variation when co-locating DNN inference jobs. It profiles each DNN, monitors microarchitectural metrics such as memory bandwidth and cache access pattern, and high-level resource utilization like CPU utilization. Based on profiling results and leveraging Modern Portfolio Theory (MPT), MediatorDNN decides on the co-location of jobs. Experimental results with various DNNs on two hardware platforms show that MediatorDNN improves throughput by up to 108% (21 % on average) compared to an approach only considering contention and ignoring resource utilization variation.

关键词： Measurement Cloud computing Microarchitecture Prevention and mitigation Artificial neural networks Throughput Hardware

来源：评论

学校读者我要写书评

暂无评论

An Optimal Scheme for WSN Based on Compressed Sensing

引用

computers, Materials & Continua 2022年第7期72卷 1053-1069页

作者： Firas Ibrahim AlZobi Ahmad Ali AlZubi Kulakov Yurii Abdullah Alharbi Jazem Mutared Alanazi Sami Smadi Department of Information Systems and Networks Faculty of Information TechnologyThe World Islamic Sciences&Education UniversityJordan Computer Science Department Community CollegeKing Saud UniversitySaudi Arabia Department of Computer Engineering National Technical University of Ukraine Igor Sikorsky Kyiv Polytechnic InstituteUkraine

Wireless sensor networks(WSNs)is one of the renowned ad hoc network technology that has vast varieties of applications such as in computer networks,bio-medical engineering,agriculture,industry and many *** has been used in the internet-of-things(IoTs)applications.A method for data collecting utilizing hybrid compressive sensing(CS)is developed in order to reduce the quantity of data transmission in the clustered sensor network and balance the network *** cluster head nodes are chosen first from each temporary cluster that is closest to the cluster centroid of the nodes,and then the cluster heads are selected in order based on the distance between the determined cluster head node and the undetermined candidate cluster head ***,each ordinary node joins the cluster that is nearest to *** greedy CS is used to compress data transmission for nodes whose data transmission volume is greater than the threshold in a data transmission tree with the Sink node as the root node and linking all cluster head *** simulation results demonstrate that when the compression ratio is set to ten,the data transfer volume is reduced by a factor of *** compared to clustering and SPT without CS,it is reduced by 75%and 65%,*** compared to SPT with Hybrid CS and Clustering with hybrid CS,it is reduced by 35%and 20%,*** and SPT without CS are compared in terms of node data transfer volume standard *** with Hybrid CS and clustering with Hybrid CS were both reduced by 62%and 80%,*** compared to SPT with hybrid CS and clustering with hybrid CS,the latter two were reduced by 41%and 19%,respectively.

关键词： Compressed sensing computer networks sensor networks ad hoc networks

来源：评论

学校读者我要写书评

暂无评论

Principled penalty-based methods for bilevel reinforcement learning and RLHF 24

Principled penalty-based methods for bilevel reinforcement l...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Han Shen Zhuoran Yang Tianyi Chen Department of Electrical Computer and Systems Engineering Rensselaer Polytechnic Institute Department of Statistics and Data Science Yale University

Bilevel optimization has been recently applied to many machine learning tasks. However, their applications have been restricted to the supervised learning setting, where static objective functions with benign structures are considered. But bilevel problems such as incentive design, inverse reinforcement learning (RL), and RL from human feedback (RLHF) are often modeled as dynamic objective functions that go beyond the simple static objective structures, which pose significant challenges of using existing bilevel solutions. To tackle this new class of bilevel problems, we introduce the first principled algorithmic framework for solving bilevel RL problems through the lens of penalty formulation. We provide theoretical studies of the problem landscape and its penalty-based (policy) gradient algorithms. We demonstrate the effectiveness of our algorithms via simulations in the Stackelberg game and RLHF.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Understanding E-consumer Online Behavior: Establishing E-commerce Performance Metrics through BERT-Light BGM Approach

Understanding E-consumer Online Behavior: Establishing E-com...

引用

2024 International Conference on Intelligent Algorithms for Computational Intelligence Systems, IACIS 2024

作者： Shaikh, Irfan Abdul Karim Akram, Nawab Gamini, Prathima Haris, A. A. Ahamed Pilli, Daniel Bhoopathy, V. Bahrain Polytechnic Faculty of Business Isa Town Bahrain Magadh Professional Institute Danapur India Sagi Rama Krishnam Raju Engineering College Department of Ece Bhimavaram India Vel Tech Rangarajan Dr.Sagunthala R&d Institute of Science and Technology Department of Cse Chennai India Vaddeswaram India Sree Rama Engineering College Department of Computer Science and Engineering Tirupathi India

ISBN: (纸本)9798350360660

This proposal draws on uses and gratifications theory to investigate how informativeness, amusement, and aggravation affect different online consumer behaviors like attitude towards the Web, Web usage, and satisfaction. The article delves into the relationship between one's attitude towards the online and its outcomes, such as web usage and contentment, as well as its causes, which include enjoyment, informativeness, and aggravation. The proposed method consists of three stages: preprocessing, feature extraction, and model training. Data cleansing and session identification are part of it. Information arrangement is another name for data pre-processing. Earth Worm Algorithm (EWA) takes advantage of a diverse population of earthworms to optimise a solution. With the help of the BERT-LightGBM, the model training procedure became more precise. The suggested method outperforms its competitors, LightGBM and BERT, with an accuracy percentage of 95.72 %. © 2024 IEEE.

关键词： Websites

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis of Various on Demand Multipath Routing Protocols 22

Performance Analysis of Various on Demand Multipath Routing ...

引用

Proceedings of the 4th International Conference on Information Management & Machine Intelligence

作者： Dinesh Goyal Kalu Ram Yadav Dept. of Computer Science & Engineering Poornima Institute of Engineering & Technology India Dept. of Computer Science & Engineering Government Polytechnic College India

ISBN: (纸本)9781450399937

MANET is a network of self-organized mobile nodes whose topology changes dynamically. There is no pre existing infrastructure is available for network setup. Nodes in MANET are mobile in nature which acts as a router as well as host. For transmitting the data and forwarding it to the desire destination MANET use multi-hop communication to transmitting the data if nodes are not in direct communication range. MANET is self configuring in nature and nodes are also work in cooperative manner to meet the performance requirements. Many on-demand multipath routing protocols have been developed in recent years, with the designers claiming that their protocol has the greatest performance. We have been done on comparison and evaluation of existing three multipath routing (SMR, ROAM and MP-DSR) algorithms using NS2 (Network Simulator 2). This paper presents performance evaluations of SMR,ROAM and MP-DSR multipath routing protocols using *** is analyzed on the average delay and packet delivery ratio metrics . For performance comparison we consider the network of 25, 50 and 100 nodes that are dispersed at random around the region of 800mX800m. Other than MP-DSR and SMR, simulation findings indicate that the ROAM protocol performs better in terms of average end-to-end delay. Result analysis also demonstrates that ROAM is also performing better in terms of PDR.

关键词： MP-DSR Proactive MANET Reactive PDR SMR Hybrid ROAM NS-2

来源：评论

学校读者我要写书评

暂无评论

MSPGait: A Mask Slice Pyramid Spatiotemporal Feature Fusion Method for Gait Recognition 25

MSPGait: A Mask Slice Pyramid Spatiotemporal Feature Fusion ...

引用

Proceedings of the 4th International Conference on computer, Artificial Intelligence and Control engineering

作者： Shaomin Lai Huimin Zhao Jian Zhong Huakang Li Kaihan Lin Huanyu Wang Wenzhe Wang School of Computer Science Guangdong Polytechnic Normal University Guangzhou Guangdong China School of Computer Science and Engineering Guangzhou Institute of Science and Technology Guangzhou Guangdong China School of Computer Science and Engineering South China University of Technology Guangzhou Guangdong China

ISBN: (纸本)9798400712647

In current gait recognition methods, researchers typically extract local features by partitioning the human body into fixed regions. However, under complex conditions, these methods lack flexibility. Additionally, traditional temporal feature extraction methods often do not fully capture the temporal information in gait data. To address these issues, we propose a mask slice pyramid spatiotemporal feature fusion gait recognition method, named MSPGait. This approach introduces three key modules: the dynamic mask slice feature module (DMSFM), the multi-scale temporal attention module (MTAM), and the pyramid adaptive feature fusion module (PAFF), which significantly enhance the model's flexibility in complex environments and its ability to model temporal features. Specifically, DMSFM improves gait feature representation and environmental adaptability by dynamically generating masks and performing slicing operations. MTAM aggregates multiscale temporal features to integrate global temporal features of body parts with short time sequence features. PAFF, on the other hand, enhances recognition accuracy and robustness by fusing spatial and temporal features at multiple levels. Comprehensive evaluations on the CASIA-B dataset validate the effectiveness and superiority of the proposed method. On the publicly available CASIA-B gait recognition dataset, our approach attains recognition accuracy rates of 98.12% for normal walking (NM), 95.82% for backpack walking (BG), and 89.59% for coat-wearing walking (CL), respectively, with an overall average recognition rate of 94.51%.

关键词： attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Anomaly-Resistant Decentralized State Estimation Under Minimum Error Entropy With Fiducial Points for Wide-Area Power Systems

引用

IEEE/CAA Journal of Automatica Sinica 2024年第1期11卷 74-87页

作者： Bogang Qu Zidong Wang Bo Shen Hongli Dong Hongjian Liu the College of Automation Engineering Shanghai University of Electric PowerShanghai 200090China IEEE the Department of Computer Science Brunel University LondonUxbridgeMiddiesex UB83PHUK the College of Information Science and Technology Donghua UniversityShanghai 200051 the Engineering Research Center of Digitalized Textile and Fashion Technology Ministry of EducationShanghai 201620China the Artificial Intelligence Energy Research Institute Northeast Petroleum UniversityDaqing 163318 the Heilongjiang Provincial Key Laboratory of Networking and Intelligent Control Northeast PetroleumDaqing 163318 the Sanya Offshore Oil&Gas Research Institute Northeast Petroleum UniversitySanya 572024China the Key Laboratory of Advanced Perception and Intelligent Control of High-End Equipment Ministry of EducationAnhui Polytechnic UniversityWuhu 2410000 the School of Mathematics and Physics Anhui Polytechnic UniversityWuhu 241000China

This paper investigates the anomaly-resistant decentralized state estimation(SE) problem for a class of wide-area power systems which are divided into several non-overlapping areas connected through transmission lines. Two classes of measurements(i.e., local measurements and edge measurements) are obtained, respectively, from the individual area and the transmission lines. A decentralized state estimator, whose performance is resistant against measurement with anomalies, is designed based on the minimum error entropy with fiducial points(MEEF) criterion. Specifically, 1) An augmented model, which incorporates the local prediction and local measurement, is developed by resorting to the unscented transformation approach and the statistical linearization approach;2) Using the augmented model, an MEEF-based cost function is designed that reflects the local prediction errors of the state and the measurement;and 3) The local estimate is first obtained by minimizing the MEEF-based cost function through a fixed-point iteration and then updated by using the edge measuring information. Finally, simulation experiments with three scenarios are carried out on the IEEE 14-bus system to illustrate the validity of the proposed anomaly-resistant decentralized SE scheme.

关键词： Decentralized state estimation(SE) measurements with anomalies minimum error entropy unscented Kalman filter wide-area power systems

来源：评论

学校读者我要写书评

暂无评论

FairCIM: Fair Interference Mitigation by DNN Switching for Latency-Sensitive Inference Jobs

FairCIM: Fair Interference Mitigation by DNN Switching for L...

引用

Autonomic Computing and Self-Organizing Systems (ACSOS),International Conference on

作者： Seyed Morteza Nabavinejad Sherief Reda Tian Guo Computer Science Department Worcester Polytechnic Institute Worcester MA School of Engineering Brown University Providence RI

ISBN: (数字)9798350363876

ISBN: (纸本)9798350363883

It is a common practice to co-locate several jobs, such as DNN inference ones, on a single machine to improve resource utilization, energy efficiency, and overall performance. However, individual job performance might degrade when these jobs compete for shared resources. This degradation can in particular affect latency-sensitive jobs bound by Service Level Agreements (SLA), and result in SLA violations. To tackle this issue and mitigate the impact of interference on latency, we design a system called FairCIM. For certain applications such as image classification, various DNNs with varying accuracy and computational complexity are proposed. Upon detection of SLA violations, FairCIM dynamically switches between those DNNs to maintain the latency in the presence of co-location interference. To ensure fairness, FairCIM selects the jobs for DNN switching based on their resource utilization. We conduct experiments using diverse DNNs, datasets, and hardware platforms to evaluate the FairCIM effectiveness. Compared to another approach that uses a resource exchange mechanism, FairCIM’s flexibility achieved by DNN switching substantially reduces SLA violations and improves the effective accuracy by up to 244%.

关键词： Accuracy Prevention and mitigation Artificial neural networks Switches Interference Hardware Resource management Faces Service level agreements Image classification

来源：评论

学校读者我要写书评

暂无评论

Enhancing Track Management Systems with Vehicle-To-Vehicle Enabled Sensor Fusion

arXiv

引用

arXiv 2024年

作者： Billington, Thomas Gwash, Ansh Kothari, Aadi Izquierdo, Lucas Talty, Timothy Department of Computer Science Virginia Polytechnic Institute State University BlacksburgVA United States Department of Electrical and Computer Engineering Virginia Polytechnic Institute State University BlacksburgVA United States

In the rapidly advancing landscape of connected and automated vehicles (CAV), the integration of Vehicle-to-Everything (V2X) communication in traditional fusion systems presents a promising avenue for enhancing vehicle perception. Addressing current limitations with vehicle sensing, this paper proposes a novel Vehicle-to-Vehicle (V2V) enabled track management system that leverages the synergy between V2V signals and detections from radar and camera sensors. The core innovation lies in the creation of independent priority track lists, consisting of fused detections validated through V2V communication. This approach enables more flexible and resilient thresholds for track management, particularly in scenarios with numerous occlusions where the tracked objects move outside the field of view of the perception sensors. The proposed system considers the implications of falsification of V2X signals which is combated through an initial vehicle identification process using detection from perception sensors. Presented are the fusion algorithm, simulated environments, and validation mechanisms. Experimental results demonstrate the improved accuracy and robustness of the proposed system in common driving scenarios, highlighting its potential to advance the reliability and efficiency of autonomous vehicles. Copyright © 2024, The Authors. All rights reserved.

关键词： Vehicle to Everything

来源：评论

学校读者我要写书评

暂无评论

Ishikawa, JESS, and Visual Analytics for engineering

Ishikawa, JESS, and Visual Analytics for Engineering

引用

2022 IEEE International Conference on Big Data, Big Data 2022

作者： Varde, Aparna S. Liang, Jianyu Yang, Zhaotong Sisson, Richard D. Montclair State University Department of Computer Science NJ United States Worcester Polytechnic Institute Mechanical and Materials Engineering MA United States

ISBN: (纸本)9781665480451

Domain-specific big data poses issues in knowledge discovery and decision support given its many Vs such as volume, variety, and visualization. In this work, we propose solutions via Ishikawa diagrams, JESS (Java Expert System Shell) with rule discovery, as well as visual analytics. We deploy data science and domain models, targeting engineering and scientific domains. Our results yield high accuracy, efficiency and cost-effectiveness. © 2022 IEEE.

关键词： Big data

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：