检索结果-内蒙古大学图书馆

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

作者： Cannelli, L. Facchinei, F. Kungurtsev, V. Scutari, G. Purdue Univ Sch Ind Engn W Lafayette IN 47907 USA Univ Roma La Sapienza Dept Comp Control & Management Engn Rome Italy Czech Tech Univ Agent Technol Ctr Prague Czech Republic

ISBN: (纸本)9781509041176

We propose a novel parallel asynchronous algorithmic framework for the minimization of the sum of a smooth (nonconvex) function and a convex (nonsmooth) regularizer. The framework hinges on Successive Convex Approximation (SCA) techniques and on a novel probabilistic model which describes in a unified way a variety of asynchronous settings in a more faithful and exhaustive way with respect to state-of-the-art models. Key features of our framework are: i) it accommodates inconsistent read, meaning that components of the variables may be written by some cores while being simultaneously read by others;ii) it covers in a unified way several existing methods;and iii) it accommodates a variety of parallel computing architectures. Almost sure convergence to stationary solutions is proved for the general case, and iteration complexity analysis is given for a specific version of our model. Numerical results show that our scheme outperforms existing asynchronous ones.

关键词： asynchronous algorithms big-data inconsistent read nonconvex constrained optimization

来源：评论

学校读者我要写书评

暂无评论

asynchronous Stochastic Convex Optimization over Random Networks: Error Bounds

Asynchronous Stochastic Convex Optimization over Random Netw...

引用

Information Theory and Applications Workshop (ITA)

作者： Touri, B. Nedic, A. Ram, S. Sundhar Univ Illinois IESE Dept Urbana IL 61801 USA Univ Illinois ECE Dept Urbana IL 61801 USA

ISBN: (纸本)9781424470143

We consider a distributed multi-agent network system where the goal is to minimize the sum of convex functions, each of which is known (with stochastic errors) to a specific network agent. We are interested in asynchronous algorithms for solving the problem over a connected network where the communications among the agent are random. At each time, a random set of agents communicate and update their information. When updating, an agent uses the (sub) gradient of its individual objective function and its own stepsize value. The algorithm is completely asynchronous as it neither requires the coordination of agent actions nor the coordination of the stepsize values. We investigate the asymptotic error bounds of the algorithm with a constant stepsize for strongly convex and just convex functions. Our error bounds capture the effects of agent stepsize choices and the structure of the agent connectivity graph. The error bound scales at best as m in the number m of agents when the agent objective functions are strongly convex.

关键词： convex optimization networked system stochastic algorithms asynchronous algorithms random consensus

来源：评论

学校读者我要写书评

暂无评论

asynchronous Byzantine-Resilient Distributed Optimization with Momentum 41

Asynchronous Byzantine-Resilient Distributed Optimization wi...

引用

第41届中国控制会议

作者： Yi Wan Yifei Qu Zuyan Zhao Shaofu Yang Chien-Shiung Wu College Southeast University School of Computer Science and Engineering Southeast University

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

In this paper,we propose a novel asynchronous Byzantine-resilient distributed optimization *** our algorithm,we adopt multiple buffers in the server to regulate the asynchronous update of *** addition,to cope with the Byzantine workers,we combine robust aggregate operator with momentum-based update,which utilize the historical information to improve the resilience of distributed *** is shown that momentum is beneficial for resisting time-coupled *** theoretically prove the convergence of our algorithm at a speed of O(1/T) in the nonconvex ***,simulation results are provided to show the effectiveness of our algorithm.

关键词： distributed optimization Byzantine learning asynchronous algorithms momentum

来源：评论

学校读者我要写书评

暂无评论

Simulations of asynchronous evolution of discrete systems

引用

Simulation Practice and Theory 1999年第4期7卷 309-324页

作者： Bahi, Jacques M. Michel, Christian J. Laboratoire de Calcul Scientifique Antenne de Belfort UMR CNRS 6623 90016 Belfort BP 527 France Lab. de Recherche en Informatique Inst. Polytech. de Sévenans 90010 Belfort Rue du Chateau a Sevenans France

Recently, a convergence theorem of asynchronous iterations of discrete dynamic systems partitioned into blocks has been proved [2]. This theorem is verified with several asynchronous block strategies. It also generalizes the chaotic iterations. Different simulations of asynchronous evolution of discrete systems performed with the research software Discrete System Evolution (DSE), lead to the first experimental results predicted by this theorem. © 1999 Elsevier Science B. V. All rights reserved.

关键词： asynchronous algorithms Boolean iterations Discrete dynamic systems Numerical simulations Research software

来源：评论

学校读者我要写书评

暂无评论

Going with the Flow: Real-Time Max-Flow on asynchronous Dynamic Graphs 6

Going with the Flow: Real-Time Max-Flow on Asynchronous Dyna...

引用

6th ACM SIGMOD Joint International Workshop on Graph Data Management Experiences and Systems / Workshop on Network Data Analytics (GRADES-NDA)

作者： Luo, Juntong Sallinen, Scott Ripeanu, Matei Univ British Columbia Vancouver BC Canada

ISBN: (纸本)9798400702013

Processing graphs that evolve over time has seen renewed attention. Processing solutions on dynamic graphs (often dubbed "graph streaming" solutions) aim to maintain the state for a graph query as the graph evolves over time, and to timely offer a solution (approximate, or precise) when requested by the user. In this space, and in the context of shared-nothing platforms, solutions have been proposed only for relatively simple problems (e.g., BFS, SSSP, PageRank), and some are limited to incremental-only evolutions traces. Support for more complex problems remains rather unexplored. To close this gap, we present a solution for the maximum flow problem that supports both add and delete events. We build this solution on top of an event-based abstraction. Integral to this abstraction is that events tied to both graph topology changes and algorithmic maintenance are processed asynchronously, concurrently, and autonomously (i.e., without shared state). We show that our implementation provides favourable time-to-solution and scales well by evaluating it on a real-world dynamic graph with 80 million edges. We compare its performance with snapshot-based solutions both internally (with our own implementation of a shared-nothing static algorithm) and externally (with Galois, a popular shared-memory framework for static graphs).

关键词： Dynamic Graphs Processing Graph Streaming asynchronous algorithms Online Graph Analytics Maximum Flow

来源：评论

学校读者我要写书评

暂无评论

An asynchronous Self-Stabilizing Maximal Independent Set Algorithm In Wireless Sensor Networks Using Two-Hop Information

An Asynchronous Self-Stabilizing Maximal Independent Set Alg...

引用

International Symposium on Networks, Computers and Communications (ISNCC)

作者： Arapoglu, Ozkan Dagdeviren, Orhan Ege Univ Int Comp Inst Izmir Turkey

ISBN: (纸本)9781728112442

Maximal independent set (MIS) has significantly important in practical applications for wireless sensor networks (WSNs). A distributed self-stabilizing system can initially start at any illegal state and takes back a legal state as long as there is no external intervention. We propose a novel distributed self-stabilizing MIS algorithm using two-hop information. It stabilizes an unstable system at most n-1 moves under an unfair distributed scheduler where n is the number of nodes in the graph. We use the message passing model as the communication model where this model is very appropriate for WSNs. The communication consumes the most energy in WSNs. So, move count is at least as important as round count where reducing the move count prolongs the lifetime of the network. We analyzed theoretically and tested it on SimPy discrete event simulator on randomly generated connected simple undirected graphs to compare with its counterparts in terms of transmitted bit count and move count against various node degrees and node counts.

关键词： self-stabilizing algorithms maximal independent set graph algorithms wireless sensor networks asynchronous algorithms

来源：评论

学校读者我要写书评

暂无评论

Master-Slave asynchronous Evolutionary Hybrid Algorithm and its Application in VANETs Routing Optimization 3

Master-Slave Asynchronous Evolutionary Hybrid Algorithm and ...

引用

3rd IIAI International Conference on Advanced Applied Informatics (IIAI-AAI)

作者： Said, Said Mohamed Nakamura, Morikazu Univ Ryukyus Dept Informat Engn Nishihara Okinawa 9030213 Japan

ISBN: (纸本)9781479941735

Hybrid algorithms incorporated with parallel processing techniques are very powerful tools for efficiently solving very complex optimization problems. We present asynchronous parallel computer architecture adaptation based on hybridization of Genetic algorithms (GAs) and Estimation of Distribution algorithms (EDAs). In this master-slave formulation, slaves perform evolutionary computation independently using GAs, while master supervises and controls the searching process. Master's role is to probabilistically study the characteristics of solution space and directs the slaves on good searching spots. This study reports some few findings on the ability of our hybrid algorithm to solve some instances of BQP problem as well as AODV routing optimization in VANETs. For both problems our hybrid algorithm has obtained best results in terms of quality of solutions as well as computational speed.

关键词： Hybrid algorithms Estimation of Distribution Algorithm Genetic algorithms Synchronous algorithms asynchronous algorithms Parallel processing Master-Slave K-means clustering BQP VANETs AODV NS-2

来源：评论

学校读者我要写书评

暂无评论

asynchronous Task-Based Parallelism in Seismic Imaging and Reservoir Modeling Simulations

Asynchronous Task-Based Parallelism in Seismic Imaging and R...

引用

作者： AlOnazi, Amani King Abdullah University of Science and Technology

学位级别：博士

The components of high-performance systems continue to become more complex on the road to exascale. This complexity is exposed at the level of: multi/many-core CPUs, accelerators (GPUs), interconnects (horizontal communication), and memory hierarchies (vertical communication). A crucial task is designing an algorithm and a programming model that scale to the same order of the HPC system size at multiple levels. This trend in HPC architecture more critically affects memory-intensive appli- cations than compute-bound applications. Accomplishing this task involves adopting less synchronous forms of the mathematical algorithm, reducing synchronization in the computational implementation, introducing more SIMT-style concurrency at the finest level of system hierarchy, and increasing arithmetic intensity as the bottleneck shifts from number of floating-point operations to number of memory accesses. This dissertation addresses these challenges in scientific simulation focusing in the dominant kernels of a memory-bound application: sparse solvers in implicit model- ing, and I/O in explicit reverse time migration in seismic imaging. We introduce asynchronous task-based parallelism into iterative algebraic preconditioners. We also introduce a task-based framework that hides the latency of I/O with computation. This dissertation targets two main applications in the oil and gas industry: reservoir simulation and seismic imaging simulation. It presents results on multi- and many- core systems and GPUs on four Top500 supercomputers: Summit, TSUBAME 3.0, Shaheen II, and Makman-2. We introduce an asynchronous implementation of four major memory-bound kernels: Algebraic multigrid (MPI+OmpSs), tridiagonal solve (MPI+OpenMP), Additive Schwarz Preconditioned Inexact Newton (MPI+MPI), and Reverse Time Migration (StarPU/StarPU+MPI and CUDA).

关键词： asynchronous algorithms Task-based runtimes MPI+X approach Task-based RTM asynchronous AMG

来源：评论

学校读者我要写书评

暂无评论

Distributed and asynchronous Algorithm for Smooth High-dimensional Function Approximation using Orthotope B-splines

Distributed and Asynchronous Algorithm for Smooth High-dimen...

引用

作者： Meyer, Johann Delft University of Technology

学位级别：硕士

Aircraft are complex systems with, in some cases, high-dimensional nonlinear interactions between control surfaces. When a failure occurs, adaptive flight control methods can be utilised to stabilise and make the aircraft controllable. Adaptive flight control methods, however, require accurate aerodynamic models - where first-order continuity is necessary for estimating the control derivatives and mitigating chattering that can reduce the longevity of components. Additionally, high-dimensional offline model identification with current approaches can take several hours for a few dimensions and this means model iterating and hyper-parameter tuning is often not feasible. Current approaches to smooth high-dimensional functional approximation are not scalable, require global communication between iteration steps, and are ill-conditioned in higher dimensions. This research develops the Distributed asynchronous B-spline (DAB) algorithm that is more robust to ill-conditioning, due to low data coverage, by using first-order methods with acceleration and weighted constraint application. This algorithm is also suitable for continuous state-spaces. Smooth aerodynamic models can be determined in exactly n·r iterations, where r is the number of continuity equations in a single dimension and n is the number of dimensions. Moreover, memory reorganisation is proposed to avoid false sharing and conflict-free use of shared memory on the GPU to ensure that the algorithm runs efficiently in parallel.

关键词： Smooth High-dimensional Linear Function Approximation Linearly-constrained Convex Quadratic Programming asynchronous algorithms Parallel algorithms Multivariate B-spline System Identification GPU Cache-aware

来源：评论

学校读者我要写书评

暂无评论

PACK: Path coloring based k-connectivity detection algorithm for wireless sensor networks

引用

AD HOC NETWORKS 2017年第Sep.期64卷 41-52页

作者： Dagdeviren, Orhan Akram, Vahid Khalilpour Ege Univ Int Comp Inst Izmir Turkey

A k-connected wireless sensor network (WSN) can tolerate failures on k-1 arbitrary nodes without loosing the connectivity between the remaining active nodes. Hence, the k value is one of the useful benchmarks that can help to measure the network reliability. Given that the nodes in a k-connected network has at least k disjoint paths to each other, we propose the path coloring based k-connectivity detection algorithm (PACK) that finds the k by counting the disjoint paths between the sink and all other nodes. The proposed algorithm has two Detection and Notification phases. In the Detection phase, all nodes find their disjoint paths to the sink and in the Notification phase the minimum detected path count, which determines the global k, is sent to the sink node. We theoretically prove that the detection range of our proposed algorithm is better than the existing distributed algorithms and uses fixed length messages with O(n Delta log(2)n) bit complexity and O(n) time complexity where n is the number of nodes and Delta is the maximum node degree. According to the comprehensive simulation results, the average correct detection ratio of proposed algorithm is more than 91% which is at least 2.3 times higher than the existing algorithms. (C) 2017 Elsevier B.V. All rights reserved.

关键词： Wireless sensor networks Network connectivity k-connectivity detection Distributed algorithms asynchronous algorithms Fault tolerance

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：