检索结果-内蒙古大学图书馆

Gossip-based distributed hierarchical algorithm for multi-cluster constrained optimisation

IET CONTROL THEORY AND APPLICATIONS 2019年第15期13卷 2346-2355页

作者： Li, Da-Kuo Shi, Chong-Xiao Yang, Guang-Hong Northeastern Univ Coll Informat Sci & Engn Shenyang 110004 Liaoning Peoples R China Northeastern Univ Minist Educ Key Lab Integrated Automat Proc Ind Shenyang 110004 Peoples R China

This study develops a gossip-based distributed hierarchical algorithm to solve the constrained convex optimisation problem over a multi-cluster network. Different from the existing works with incremental communication approach, a gossip protocol is applied to achieve inter-cluster information exchange. At each iteration, only two neighbour clusters are randomly woken up to exchange their new updates while others keep their latest information, which relaxes the restrictions on the inter-cluster communication topology. Technically, to investigate the intra-cluster consensus, a new analysing approach is exploited and some different techniques including the non-expansiveness of projection operator and the supermartingale convergence results are employed, based on which it is proved that estimates of all agents in the multi-cluster network jointly converge to the optimal solution with probability one. A numerical example on constrained lasso regression is provided to validate the theoretical results.

关键词： protocols regression analysis convex programming optimisation graph theory distributed algorithms probability wireless sensor networks pattern clustering convergence gossip-based hierarchical algorithm multicluster constrained optimisation constrained convex optimisation problem multicluster network existing works incremental communication approach gossip protocol inter-cluster information exchange neighbour clusters latest information inter-cluster communication topology intra-cluster consensus analysing approach different techniques constrained lasso regression

来源：评论

学校读者我要写书评

暂无评论

distributed CONGESTBC constant approximation of MDS in bounded genus graphs

引用

THEORETICAL COMPUTER SCIENCE 2019年 757卷 1-10页

作者： Czygrinow, Andrzej Hanckowiak, Michal Wawrzyniak, Wojciech Witkowski, Marcin Arizona State Univ Sch Math & Stat Sci Tempe AZ 85287 USA Adam Mickiewicz Univ Fac Math & Comp Sci Poznan Poland

We give a distributed algorithm which finds a constant approximation of a minimum dominating set in graphs of constant genus. The algorithm works in the CONGEST(BC) model (distributed, synchronous, with short messages and broadcast type of transmissions) and runs in a constant number of distributed rounds. (C) 2018 Elsevier B.V. All rights reserved.

关键词： distributed algorithms Minimum dominating set Bounded genus graphs Local algorithms

来源：评论

学校读者我要写书评

暂无评论

distributed Power Control for Large Energy Harvesting Networks: A Multi-Agent Deep Reinforcement Learning Approach

引用

IEEE TRANSACTIONS ON COGNITIVE COMMUNICATIONS AND NETWORKING 2019年第4期5卷 1140-1154页

作者： Sharma, Mohit K. Zappone, Alessio Assaad, Mohamad Debbah, Merouane Vassilaras, Spyridon Univ Paris Saclay Cent Supelec F-91192 Gif Sur Yvette France Cent Supelec Lab Signaux & Syst F-91190 Gif Sur Yvette France Univ Cassino & Southern Lazio Dept Elect & Informat Engn I-03043 Cassino Italy Huawei France R&D Math & Algorithm Sci Lab F-92100 Paris France

In this paper, we develop a multi-agent reinforcement learning (MARL) framework to obtain online power control policies for a large energy harvesting (EH) multiple access channel, when only causal information about the EH process and wireless channel is available. In the proposed framework, we model the online power control problem as a discrete-time mean-field game (MFG), and analytically show that the MFG has a unique stationary solution. Next, we leverage the fictitious play property of the mean-field games, and the deep reinforcement learning technique to learn the stationary solution of the game, in a completely distributed fashion. We analytically show that the proposed procedure converges to the unique stationary solution of the MFG. This, in turn, ensures that the optimal policies can be learned in a completely distributed fashion. In order to benchmark the performance of the distributed policies, we also develop a deep neural network (DNN) based centralized as well as distributed online power control schemes. Our simulation results show the efficacy of the proposed power control policies. In particular, the DNN based centralized power control policies provide a very good performance for large EH networks for which the design of optimal policies is intractable using the conventional methods such as Markov decision processes. Further, performance of both the distributed policies is close to the throughput achieved by the centralized policies.

关键词： Artificial neural networks distributed algorithms energy harvesting learning (artificial intelligence) multi-agent systems multiple access

来源：评论

学校读者我要写书评

暂无评论

Asynchronous Bayesian Learning over a Network

Asynchronous Bayesian Learning over a Network

引用

IEEE Conference on Decision and Control

作者： Kinjal Bhar He Bai Jemin George Carl Busart Oklahoma State University Stillwater OK USA U.S. Army Research Laboratory Adelphi MD USA

ISBN: (数字)9781665467612

ISBN: (纸本)9781665467629

We present a practical asynchronous data fusion model for networked agents to perform distributed Bayesian learning without sharing raw data. Our algorithm employs unadjusted Langevin dynamics with a gossip-based protocol for sampling, coupled with an event-triggered mechanism to further reduce communication between gossiping agents. These mechanisms drastically reduce communication overhead and help avoid bottlenecks commonly experienced with distributed algorithms. In addition, the algorithm is expected to increase resilience to occasional link failure. We establish mathematical guarantees for our algorithm and demonstrate its effectiveness via a numerical experiment.

关键词： Protocols Heuristic algorithms distributed databases Data integration Data models Bayes methods distributed algorithms

来源：评论

学校读者我要写书评

暂无评论

Multi-Agent System Based distributed Pattern Search Algorithm for Non-Convex Economic Load Dispatch in Smart Grid

引用

IEEE TRANSACTIONS ON POWER SYSTEMS 2019年第3期34卷 2093-2102页

作者： Li, Fangyuan Qin, Jiahu Kang, Yu Univ Sci & Technol China Dept Automat Hefei 230027 Anhui Peoples R China Univ Sci & Technol China Dept Automat State Key Lab Fire Sci Hefei 230027 Anhui Peoples R China Univ Sci & Technol China Inst Adv Technol Hefei 230027 Anhui Peoples R China Chinese Acad Sci Key Lab Technol Geospatial Informat Proc & Applic Beijing 100190 Peoples R China

Since generation units usually subject to valve point effects and there is a trend of ever-growing number of distributed energy sources, distributed optimization algorithms for non-convex economic load dispatch are more desirable than the centralized ones. In this paper, a distributed pattern search algorithm (DPSA) for non-convex economic dispatch is developed. To develop such an algorithm, a flooding-based topology discovery algorithm (FBTDA) is proposed first. Theoretic analysis shows that the FBTDA algorithm converges in the least number of iterations compared with existing methods. Then, the DPSA algorithm is developed by integrating the original pattern search algorithm and the FBTDA algorithm under the framework of multi-agent systems. The DPSA algorithm has a rather weak requirement on the communication topology. Rigorous convergence analysis is provided as opposed to previous literatures where convergence property is often not investigated. The correctness and efficiency of the proposed algorithms are verified by numerical simulations, which also show that the DPSA algorithm can improve performance by actively choosing communication topology and integrating other global search methods. Comparisons with traditional methods show that besides the potential benefits arising from the absence of central nodes, the DPSA algorithm has comparable performance to these methods.

关键词： Convergence distributed algorithms multi-agent systems power generation dispatch

来源：评论

学校读者我要写书评

暂无评论

distributed multi-vehicle task assignment in a time-invariant drift field with obstacles

引用

IET CONTROL THEORY AND APPLICATIONS 2019年第17期13卷 2886-2893页

作者： Bai, Xiaoshan Yan, Weisheng Cao, Ming Xue, Dong Univ Groningen Fac Sci & Engn NL-9747 AG Groningen Netherlands Northwestern Polytech Univ Sch Marine Sci & Technol 127 West Youyi Rd Xian 710072 Shaanxi Peoples R China Tech Univ Munich Dept Elect & Comp Engn Arcisstr 21 D-80209 Munich Germany

This study investigates the task assignment problem where a fleet of dispersed vehicles needs to visit multiple target locations in a time-invariant drift field with obstacles while trying to minimise the vehicles' total travel time. The vehicles have different capabilities, and each kind of vehicles can visit a certain type of the target locations;each target location might require to be visited more than once by different kinds of vehicles. The task assignment problem has been proven to be NP-hard. A path planning algorithm is first designed to minimise the time for a vehicle to travel between two given locations through the drift field while avoiding any obstacle. The path planning algorithm provides the travel cost matrix for the target assignment, and generates routes once the target locations are assigned to the vehicles. Then, a distributed algorithm is proposed to assign the target locations to the vehicles using only local communication. The algorithm guarantees that all the visiting demands of every target will be satisfied within a total travel time that is at worst twice of the optimal when the travel cost matrix is symmetric. Numerical simulations show that the algorithm can lead to solutions close to the optimal.

关键词： remotely operated vehicles mobile robots distributed algorithms computational complexity minimisation total travel time travel cost matrix multivehicle task assignment time-invariant drift field task assignment problem dispersed vehicles multiple target locations path planning algorithm target assignment distributed algorithm numerical simulation

来源：评论

学校读者我要写书评

暂无评论

Efficient and Scalable Functional Dependency Discovery on distributed Data-Parallel Platforms

引用

IEEE TRANSACTIONS ON PARALLEL AND distributed SYSTEMS 2019年第12期30卷 2663-2676页

作者： Zhu, Guanghui Wang, Qian Tang, Qiwei Gu, Rong Yuan, Chunfeng Huang, Yihua Nanjing Univ Natl Key Lab Novel Software Technol Nanjing 210008 Jiangsu Peoples R China

Functional dependencies (FDs) play a very important role in many data management tasks such as schema normalization, data cleaning, and query optimization. Meanwhile, there are ever-increasing application demands for efficient FD discovery on large-scale datasets. Unfortunately, due to huge runtime and memory overhead, the existing single-machine FD discovery algorithms are inefficient for large-scale datasets. Recently, distributed data-parallel computing has become the de facto standard for large-scale data processing. However, it is challenging to design an efficient distributed FD discovery algorithm. In this paper, we present SmartFD, which is an efficient and scalable algorithm for distributed FD discovery. First, we propose a novel attribute sorting-based algorithm framework. Next, to discover all the FDs grouped by a given attribute, we propose an efficient distributed algorithm Attribute-centric Functional Dependency Discovery (AFDD). In AFDD, we design an Fast Sampling and Early Aggregation (FSEA) mechanism to improve the efficiency of distributed sampling and propose a memory-efficient index-based method for distributed FD validation. Moreover, AFDD employs an attribute-parallel method to accelerate the pruning-and-generation of candidate FDs. Furthermore, we propose an adaptive switching strategy between distributed sampling and distributed validation based on the unified time-based efficiency metric. Also, we employ a distributed probing based method to make the switching strategy more accurate. Experimental results on Apache Spark reveal that SmartFD outperforms the state-of-the-art single-machine algorithm HyFD and the existing distributed algorithm HFDD with 3.2 & x00D7;-44.9 & x00D7;and 2.5 & x00D7;-455.7 & x00D7;speedup respectively. Moreover, SmartFD achieves good row scalability and column scalability. Additionally, SmartFD has sub-linear node scalability.

关键词： distributed databases Scalability Remuneration Lattices distributed algorithms Switches Query processing Functional dependency discovery distributed computing data-parallel algorithms

来源：评论

学校读者我要写书评

暂无评论

distributed Newton Method for Large-Scale Consensus Optimization

引用

IEEE TRANSACTIONS ON AUTOMATIC CONTROL 2019年第10期64卷 3983-3994页

作者： Tutunov, Rasul Bou-Ammar, Haitham Jadbabaie, Ali Huawei Technol Res & Dev Shenzhen Peoples R China MIT Inst Data Syst & Soc 77 Massachusetts Ave Cambridge MA 02139 USA

In this paper, we propose a distributed Newton method for decenteralized optimization of large sums of convex functions. Our proposed method is based on creating a set of separable finite sum minimization problems by utilizing a decomposition technique known as Global Consensus that distributes the computation across nodes of a graph and enforces a consensus constraint among the separated variables. The key idea is to exploit the sparsity of the dual Hessian and recast the computation of the Newton step as one of the efficiently solving symmetric diagonally dominant linear equations. We show that our method outperforms the state-of-the-art algorithms, including ADMM. We validate our algorithm both theoretically and empirically. On the theory side, we demonstrate that our algorithm exhibits superlinear convergence within a neighborhood of optimality. Empirically, we show the superiority of this new method on a variety of large-scale optimization problems. The proposed approach is scalable to large problems and has a low communication overhead.

关键词： Convergence distributed algorithms

来源：评论

学校读者我要写书评

暂无评论

A Real-Time and Fully distributed Approach to Motion Planning for Multirobot Systems

引用

IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS 2019年第12期49卷 2636-2650页

作者： Zhou, Yuan Hu, Hesuan Liu, Yang Lin, Shang-Wei Ding, Zuohua Nanyang Technol Univ Sch Comp Sci & Engn Coll Engn Singapore 639798 Singapore Xidian Univ Sch Electromech Engn Xian 710071 Shaanxi Peoples R China Xi An Jiao Tong Univ State Key Lab Mfg Syst Engn Xian 710054 Shaanxi Peoples R China Zhejiang Sci Tech Univ Sch Informat Sci Hangzhou 310018 Zhejiang Peoples R China

Motion planning is one of the most critical problems in multirobot systems. The basic target is to generate a collision-free trajectory for each robot from its initial position to the target position. In this paper, we study the trajectory planning for the multirobot systems operating in unstructured and changing environments. Each robot is equipped with some sensors of limited sensing ranges. We propose a fully distributed approach to planning trajectories for such systems. It combines the model predictive control (MPC) strategy and the incremental sequential convex programming (iSCP) method. The MPC framework is applied to detect the local running environment real-timely with the concept of receding horizon. For each robot, a nonlinear programming is built in its current prediction horizon. To construct its own optimization problem, a robot first needs to communicate with its neighbors to retrieve their current states. Then, the robot predicts the neighbors' future positions in the current horizon and constructs the problem without waiting for the prediction information from its neighbors. At last, each robot solves its problem independently via the iSCP method such that the robot can move autonomously. The proposed method is polynomial in its computational complexity.

关键词： distributed algorithms motion planning multirobot systems trajectory optimization

来源：评论

学校读者我要写书评

暂无评论

distributed Clustering of Text Collections

引用

IEEE ACCESS 2019年 7卷 155671-155685页

作者： Zamora, Juan Allende-Cid, Hector Mendoza, Marcelo Pontificia Univ Catolica Valparaiso Inst Estadist Valparaiso 2340023 Chile Pontificia Univ Catolica Valparaiso Escuela Ingn Informat Valparaiso 2362807 Chile Univ Tecn Federico Santa Maria Ctr Cient & Tecnol Valparaiso Valparaiso 2390123 Chile

Current data processing tasks require efficient approaches capable of dealing with large databases. A promising strategy consists in distributing the data along with several computers that partially solve the undertaken problem. Finally, these partial answers are integrated to obtain a final solution. We introduce distributed shared nearest neighbors (D-SNN), a novel clustering algorithm that work with disjoint partitions of data. Our algorithm produces a global clustering solution that achieves a competitive performance regarding centralized approaches. The algorithm works effectively with high dimensional data, being advisable for document clustering tasks. Experimental results over five data sets show that our proposal is competitive in terms of quality performance measures when compared to state of the art methods.

关键词： Clustering algorithms distributed databases Partitioning algorithms Task analysis Computers Parallel processing Approximation algorithms distributed algorithms distributed text clustering high dimensional 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：