检索结果-内蒙古大学图书馆

NOMAD: Non-locking, stOchastic Multi-machine algorithm for Asynchronous and decentralized matrix completion

PROCEEDINGS OF THE VLDB ENDOWMENT 2014年第11期7卷 975-986页

作者： Yun, Hyokun Yu, Hsiang-Fu Hsieh, Cho-Jui Vishwanathan, S. V. N. Dhillon, Inderjit Purdue Univ W Lafayette IN 47907 USA Univ Texas Austin Austin TX 78712 USA

We develop an efficient parallel distributed algorithm for matrix completion, named NOMAD (Non-locking, stOchastic Multi-machine algorithm for Asynchronous and decentralized matrix completion). NOMAD is a decentralized algorithm with non-blocking communication between processors. One of the key features of NOMAD is that the ownership of a variable is asynchronously transferred between processors in a decentralized fashion. As a consequence it is a lock-free parallel algorithm. In spite of being asynchronous, the variable updates of NOMAD are serializable, that is, there is an equivalent update ordering in a serial implementation. NOMAD outperforms synchronous algorithms which require explicit bulk synchronization after every iteration: our extensive empirical evaluation shows that not only does our algorithm perform well in distributed setting on commodity hardware, but also outperforms stateof-the-art algorithms on a HPC cluster both in multi-core and distributed memory settings.

关键词： Iterative methods Locks (fasteners) Stochastic systems Commodity hardware decentralized algorithms Distributed Memory Empirical evaluations Matrix completion Parallel distributed algorithms State of the art algorithms Synchronous algorithms

来源：评论

学校读者我要写书评

暂无评论

Assessing operator strategies for real-time replanning of multiple unmanned vehicles

引用

INTELLIGENT DECISION TECHNOLOGIES-NETHERLANDS 2012年第3期6卷 221-231页

作者： Clare, Andrew S. Maere, Pierre C. P. Cummings, M. L. MIT Dept Aeronaut & Astronaut Cambridge MA 02139 USA Delft Univ Technol Delft Netherlands

Future unmanned vehicles systems will invert the operator-to-vehicle ratio so that one operator controls a decentralized network of heterogeneous unmanned vehicles. This study examines the impact of allowing an operator to adjust the rate of prompts to view automation-generated plans on system performance and operator workload. Results showed that the majority of operators chose to adjust the replan prompting rate. The initial replan prompting rate had a significant framing effect on the replan prompting rates chosen throughout a scenario. Higher initial replan prompting rates led to significantly lower system performance. Operators successfully self-regulated their task-switching behavior to moderate their workload.

关键词： Unmanned vehicles automation decentralized algorithms human-machine interface human supervisory control human-computer interaction mental workload

来源：评论

学校读者我要写书评

暂无评论

S-ADDOPT: decentralized Stochastic First-Order Optimization Over Directed Graphs

引用

IEEE CONTROL SYSTEMS LETTERS 2021年第3期5卷 953-958页

作者： Qureshi, Muhammad, I Xin, Ran Kar, Soummya Khan, Usman A. Tufts Univ Dept Elect Engn Medford MA 02155 USA Carnegie Mellon Univ Dept Elect & Comp Engn Pittsburgh PA 15232 USA

In this letter, we study decentralized stochastic optimization to minimize a sum of smooth and strongly convex cost functions when the functions are distributed over a directed network of nodes. In contrast to the existing work, we use gradient tracking to improve certain aspects of the resulting algorithm. In particular, we propose the S-ADDOPT algorithm that assumes a stochastic first-order oracle at each node and show that for a constant step-size alpha, each node converges linearly inside an error ball around the optimal solution, the size of which is controlled by alpha. For decaying step-sizes O(1/k), we show that S-ADDOPT reaches the exact solution sublinearly at O(1/k) and its convergence is asymptotically network-independent. Thus the asymptotic behavior of S-ADDOPT is comparable to the centralized stochastic gradient descent. Numerical experiments over both strongly convex and non-convex problems illustrate the convergence behavior and the performance comparison of the proposed algorithm.

关键词： Stochastic optimization first-order methods decentralized algorithms directed graphs neural networks

来源：评论

学校读者我要写书评

暂无评论

Agent-Supervisor Coordination for decentralized Event-Triggered Optimization

引用

IEEE CONTROL SYSTEMS LETTERS 2022年 6卷 1970-1975页

作者： Srivastava, Priyank Cavraro, Guido Cortes, Jorge MIT Dept Mech Engn Cambridge MA 02139 USA Natl Renewable Energy Lab Power Syst Engn Ctr Golden CO 80401 USA Univ Calif San Diego Dept Mech & Aerosp Engn La Jolla CA 92093 USA

This letter proposes decentralized resource-aware coordination schemes for solving network optimization problems defined by objective functions that combine locally evaluable costs with network-wide coupling components. These methods are well suited for a group of supervised agents trying to solve an optimization problem under mild coordination requirements. Each agent has information on its local cost and coordinates with the network supervisor for information about the coupling term of the cost. The proposed approach is feedback-based and asynchronous by design, guarantees anytime feasibility, and ensures the asymptotic convergence of the network state to the desired optimizer. Numerical simulations on a power system example illustrate our results.

关键词： Couplings Costs Heuristic algorithms Convergence Linear programming Power system dynamics Government Event-triggered control optimization decentralized algorithms

来源：评论

学校读者我要写书评

暂无评论

Informationally decentralized System Resource Management for Multiple Multimedia Tasks

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2009年第9期19卷 1352-1364页

作者： Foo, Brian van der Schaar, Mihaela Univ Calif Los Angeles Dept Elect Engn Los Angeles CA 90095 USA

An increasing number of networked and battery-powered devices require the simultaneous processing of multiple high-complexity dynamic multimedia tasks. Existing resource allocation solutions include both centralized approaches and distributed approaches. Centralized approaches typically have two fundamental limitations. First, multimedia algorithms typically have a large number of operating points and thus requires high implementation complexity to optimally allocate resources across multiple applications, making centralized solutions unsuitable for dynamic environments. Second, autonomous multimedia applications may not be willing to reveal information about their private utility functions, since such information can be exploited by other selfish applications to gain an unfair performance advantage by strategically misusing the system resources. On the other hand, existing decentralized solutions typically solve only resource-constrained utility maximization problems, without regard to other system platform-dependent costs and objectives, and without considering application dynamics. To address these limitations, we present in this paper a decentralized low-complexity resource management solution that can optimize a variety of system performance objectives without requiring applications to provide their utilities. We demonstrate analytically and experimentally that our algorithms converge quickly to their respective optimal solutions and are thus practical in dynamic environments.

关键词： Autonomous multimedia applications decentralized algorithms information privacy multitask resource management resource-aware system optimization

来源：评论

学校读者我要写书评

暂无评论

decentralized Reconfigurable Multi-Robot Coordination from Local Connectivity and Collision Avoidance Specifications

引用

IFAC-PapersOnLine 2017年第1期50卷 15798-15803页

作者： Vrohidis, Constantinos Bechlioulis, Charalampos P. Kyriakopoulos, Kostas J. Control Systems Laboratory School of Mechanical Engineering National Technical University of Athens 9 Heroon Polytechniou Street Zografou15780 Greece

Multi-robot motion planning from local connectivity specifications in unknown workspaces necessitates for reconfigurability, since real-time obstacle detection is likely to raise contradictions between collision avoidance and local connectivity specifications. Thus, a key issue in control synthesis is to allow topological changes in the communication graph in case of infeasibility, while maintaining the global connectivity of the robotic network and guaranteeing collision-free motion. Towards this direction, we propose: i) a distributed algorithm that coordinates robots, whenever feasible, based on local connectivity as well as collision avoidance specifications and ii) a novel decentralized algorithm that dynamically reconfigures the communication topology in case of conflicts with the collision avoidance specifications, without compromising the global connectivity of the communication graph. In the continuous control system, provably correct collision avoidance and robust synchronized motion are secured, employing the prescribed performance control methodology that ensures predefined transient and steady state response. Finally, simulated examples clarify the proposed methodology and verify its efficiency. © 2017

关键词： Robot programming Collision avoidance Industrial robots Motion planning Multi agent systems Multipurpose robots Obstacle detectors Specifications Topology Communication topologies decentralized algorithms Distributed coordination Multi robot coordination Multi robot motion planning Prescribed performance Steady state response Task specifications

来源：评论

学校读者我要写书评

暂无评论

Identifying Surrounds and Engulfs Relations in Mobile and Coordinate-Free Geosensor Networks

引用

ACM TRANSACTIONS ON SPATIAL algorithms AND SYSTEMS 2018年第2期4卷 1–21页

作者： Both, Alan Duckham, Matt Worboys, Michael F. RMIT Univ GPO Box 2476 Melbourne Vic 3001 Australia Univ Greenwich London England Old Royal Naval Coll London SE10 9LS England

This article concerns the definition and identification of qualitative spatial relationships for the full and partial enclosure of spatial regions. The article precisely defines three relationships between regions-"surrounds," "engulfs," and "envelops"-highlighting the correspondence to similar definitions in the literature. An efficient algorithm capable of identifying these qualitative spatial relations in a network of dynamic (mobile) geosensor nodes is developed and tested. The algorithms are wholly decentralized, and operate in-network with no centralized control. The algorithms are also " coordinate-free," able to operate in distributed spatial computing environments where coordinate locations are expensive to capture or otherwise unavailable. Experimental evaluation of the algorithms designed demonstrates the efficiency of the approach. Although the algorithm communication complexity is dominated by an overall worst-case O(n(2)) leader election algorithm, the experiments show in practice an average-case complexity approaching linear, O(n(1.1)).

关键词： Geosensor networks qualitative spatial reasoning moving objects Voronoi regions decentralized algorithms

来源：评论

学校读者我要写书评

暂无评论

A decentralized Clustering Scheme for Dynamic Downlink Base Station Cooperation

引用

IEICE TRANSACTIONS ON COMMUNICATIONS 2010年第12期E93B卷 3656-3659页

作者： Zhou, Sheng Gong, Jie Jia, Yunjian Niu, Zhisheng Tsinghua Univ Tsinghua Natl Lab Informat Sci & Technol Beijing 100084 Peoples R China Hitachi Ltd Cent Res Lab Kokubunji Tokyo 1858601 Japan

Base station (BS) cooperation is a promising technique to suppress co channel interference for cellular networks However practical limitations constrain the scale of cooperation thus the network is divided into small disjoint BS cooperation groups namely clusters A decentralized scheme for BS cluster formation is proposed based on efficient BS negotiations of which the feedback overhead per user is nearly Irrelevant to the network size and the number of iteration rounds scales very slowly with the network size Simulations show that our decentralized scheme provides significant sum rate gain over static clustering and performs almost the same as the existing centralized approach The proposed scheme is well suited for large scale cellular networks due to its low overhead and complexity

关键词： interference mitigation base station cooperation decentralized algorithms

来源：评论

学校读者我要写书评

暂无评论

SYNCHRONIZED DISTRIBUTED TERMINATION

引用

IEEE TRANSACTIONS ON SOFTWARE ENGINEERING 1985年第10期11卷 1136-1140页

作者： SZYMANSKI, B SHI, Y PRYWES, NS UNIV PENN MOORE SCH ELECT ENGNDEPT COMP & INFORMAT SCIPHILADELPHIAPA 19104 TEMPLE UNIV PHILADELPHIAPA 19122

The distributed termination problem involves determining the instant at which a distributed computation, which is accessed in common by a number of processes within a network, can be terminated. Each process can only determine when its own local termination conditions have been met; an algorithm is required to detect when local termination conditions have been met by all processes. The present paper develops an efficient synchronized distributed termination algorithm that does not require prior knowledge of network configuration or number of processes. The network is assumed to be strongly connected with no central controller. Unidirectional channels connect distributed processes. Once all processes have satisfied local termination conditions, distributed computation termination will require a number of steps proportional to the network diameter.

关键词： Termination Detection decentralized algorithms Distributed Programming Distributed Systems Networks

来源：评论

学校读者我要写书评

暂无评论

Push-SAGA: A decentralized Stochastic Algorithm With Variance Reduction Over Directed Graphs

引用

IEEE CONTROL SYSTEMS LETTERS 2022年 6卷 1202-1207页

作者： Qureshi, Muhammad I. Xin, Ran Kar, Soummya Khan, Usman A. Carnegie Mellon Univ Dept Elect & Comp Engn Pittsburgh PA 15213 USA Tufts Univ Dept Elect Engn Medford MA 02155 USA Tufts Univ Dept Elect & Comp Engn Medford MA 02155 USA

In this letter, we propose Push-SAGA, a decentralized stochastic first-order method for finite-sum minimization over a directed network of nodes. Push-SAGA combines node-level variance reduction to remove the uncertainty caused by stochastic gradients, network-level gradient tracking to address the distributed nature of the data, and push-sum consensus to tackle directed information exchange. We show that Push-SAGA achieves linear convergence to the exact solution for smooth and strongly convex problems and is thus the first linearly-convergent stochastic algorithm over arbitrary strongly connected directed graphs. We also characterize the regime in which Push-SAGA achieves a linear speed-up compared to its centralized counterpart and achieves a network-independent convergence rate. We illustrate the behavior and convergence properties of Push-SAGA with the help of numerical experiments on strongly convex and non-convex problems.

关键词： Convergence Directed graphs Radio frequency Uncertainty Steady-state Robots Minimization Stochastic optimization variance reduction first-order methods decentralized algorithms directed graphs

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：