检索结果-内蒙古大学图书馆

decentralized GRADIENT DESCENT MAXIMIZATION METHOD FOR COMPOSITE NONCONVEX STRONGLY-CONCAVE MINIMAX PROBLEMS

SIAM JOURNAL ON OPTIMIZATION 2024年第1期34卷 1006-1044页

作者： Xu, Yangyang Rensselaer Polytech Inst Dept Math Sci Troy NY 12180 USA

Minimax problems have recently attracted a lot of research interests. A few efforts have been made to solve decentralized nonconvex strongly -concave (NCSC) minimax-structured optimization;however, all of them focus on smooth problems with at most a constraint on the maximization variable. In this paper, we make the first attempt on solving composite NCSC minimax problems that can have convex nonsmooth terms on both minimization and maximization variables. Our algorithm is designed based on a novel reformulation of the decentralized minimax problem that introduces a multiplier to absorb the dual consensus constraint. The removal of dual consensus constraint enables the most aggressive (i.e., local maximization instead of a gradient ascent step) dual update that leads to the benefit of taking a larger primal stepsize and better complexity results. In addition, the decoupling of the nonsmoothness and consensus on the dual variable eases the analysis of a decentralized algorithm;thus our reformulation creates a new way for interested researchers to design new (and possibly more efficient) decentralized methods on solving NCSC minimax problems. We show a global convergence result of the proposed algorithm and an iteration complexity result to produce a (near) stationary point of the reformulation. Moreover, a relation is established between the (near) stationarities of the reformulation and the original formulation. With this relation, we show that when the dual regularizer is smooth, our algorithm can have lower complexity results (with reduced dependence on a condition number) than existing ones to produce a near -stationary point of the original formulation. Numerical experiments are conducted on a distributionally robust logistic regression to demonstrate the performance of the proposed algorithm.

关键词： composite minimax problem nonconvex strongly-concave decentralized algorithm iteration complexity

来源：评论

学校读者我要写书评

暂无评论

Linear Precoding to Optimize Throughput, Power Consumption and Energy Efficiency in MIMO Wireless Sensor Networks

引用

IEEE TRANSACTIONS ON COMMUNICATIONS 2018年第5期66卷 2122-2136页

作者： Liu, Yang Li, Jing Lehigh Univ Elect & Comp Engn Dept Bethlehem PA 18015 USA

This paper considers joint precoder design to optimize throughput, power consumption and energy efficiency (EE) in the context of multi-antenna wireless sensor networks with coherent multiple access channels. To maximize throughput, both centralized and decentralized algorithms are developed. Our centralized algorithm obtains a new second order cone programming formulation of the problem, which is different from related works and can apply to more generic system setup compared to existing literature. In addition, noting the fact that all existing solutions in literature are centralized based, we propose a novel decentralized solution and analyses its convergence. Besides the throughput maximization, the power consumption and EE problems are also attacked. To optimize these two metrics, a decentralized algorithm based on dual-decomposition and block successive upper-bound method has been developed, which runs in parallel with semi-analytical solutions and has provable strong convergence. A sufficient condition for the validity of the decentralized method is obtained. Extensive numerical results are presented to consolidate our findings.

关键词： Wireless sensor network throughput energy efficiency (EE) power consumption decentralized algorithm

来源：评论

学校读者我要写书评

暂无评论

Delay-Constrained Energy Optimization in High-Latency Sensor Networks

引用

IEEE SENSORS JOURNAL 2017年第13期17卷 4287-4298页

作者： Ponnavaikko, Poongovan Wilson, Sarah Kate Stojanovic, Milica Holliday, JoAnne Yassin, Kamal Santa Clara Univ Comp Engn Dept Santa Clara CA 95053 USA Santa Clara Univ Elect Engn Dept Santa Clara CA 95053 USA Northeastern Univ Elect & Comp Engn Dept Boston MA 02115 USA ASSIA Inc Redwood City CA 94065 USA

Sensor networks deployed in high-latency environments, such as underwater acoustic and satellite channels, find critical applications in disaster prevention and tactical surveillance. The sensors in these networks have limited energy reserves. In order to extend the lifetime of these sensors, energy must be conserved in all layers of the protocol stack. In addition to long propagation delays, these channels are characterized by limited bandwidth and a lack of well-established closed-form analytical models. This fact makes finding cross-layer energy-optimal solutions a difficult problem to solve. Our goal is to compute near-optimal routes, schedules and transmit power levels for delay-constrained applications of high-latency sensor networks. We present a mixed-integer programming relaxation of the optimization problem. We further propose a decentralized algorithm to iteratively solve the relaxed optimization problem. Comparative simulation analysis shows that our decentralized approach is approximately 3 similar to 6 dB more energy-efficient and 2 similar to 5 dB more throughput-efficient than the heuristic, time-sensitive greedy forwarding, and least- cost routing algorithms.

关键词： Cross-layer energy optimization high-latency sensor networks underwater acoustic medium delay-intolerant applications large propagation delays decentralized algorithm

来源：评论

学校读者我要写书评

暂无评论

Robust Linear Beamforming in Wireless Sensor Networks

引用

IEEE TRANSACTIONS ON COMMUNICATIONS 2019年第6期67卷 4450-4463页

作者： Liu, Yang Li, Jing Wang, Hao Lehigh Univ Elect & Comp Engn Dept Bethlehem PA 18015 USA ShanghaiTech Univ Sch Informat Sci & Technol Shanghai 201210 Peoples R China

A typical wireless sensor network (WSN) has multiple sensors, each of which obtains noisy observation of one common physical event and sends the observation to a fusion center (FC) for post-processing. This paper provides a comprehensive research on robust linear transceiver design in the presence of imperfect channel-state information (CSI) in multi-input multi-output (MIMO) WSNs. For the CSI uncertainty, the classical ellipsoid model is assumed, where channel estimation errors live in an ellipsoid. Two robust beamforming problems are considered: 1) the worst-case mean-square-error (MSE) minimization with limited power and 2) sum-power minimization with guaranteed worst-case MSE. For both problems, centralized and decentralized solutions are developed. Moreover, an interesting fundamental relation between the metrics of signal-to-noise ratio (SNR) and MSE has been presented, which extends our proposed solutions to handle the transceiver design when SNR is considered. Extensive numerical results are presented to confirm our findings.

关键词： Wireless sensor network (WSN) robust linear beamforming alternative direction method of multipliers (ADMM) decentralized algorithm

来源：评论

学校读者我要写书评

暂无评论

A general scalable and accurate decentralized level monitoring method for large-scale dynamic service provision in hybrid clouds

引用

FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE 2013年第5期29卷 1235-1253页

作者： Fu, Yongquan Wang, Yijie Biersack, Ernst Natl Univ Def Technol Coll Comp Sci Natl Key Lab Parallel & Distributed Proc Changsha 410073 Hunan Peoples R China

Hybrid cloud computing combines private clouds with geographically-distributed resources from public clouds, desktop grids or in-house gateways to provide the most flexibility of each kind of cloud platforms. Service provisioning for wide-area applications such as cloud backup or cloud network games is sensitive to wide-area network metrics such as round trip time, bandwidth, or loss rates. In order to optimize the quality of the service provision in hybrid clouds, it is highly valuable for the hybrid clouds to collect detailed network metrics between participating nodes of the hybrid clouds. However, since nodes can be large-scale and dynamic, the network metrics may be diverse for different cloud services, it is challenging to increase the generality, scalability, accuracy, and the robustness of the measurement process. We propose a novel distributed level monitoring method HPM (Hierarchical Performance Measurement) satisfying these requirements. For each kind of network metric, HPM represents the degree of pairwise closeness with discrete level values inspired by the hierarchical clustering tree. HPM maps probed metric to discrete levels based on an existing distributed K-means clustering method that helps maximize the similarity of the network metric in the same level, which therefore optimizes the matching between pairwise levels and the real-world pairwise proximity. Furthermore, for scalability reasons, HPM computes the pairwise levels with decentralized coordinates. Each node independently maintains its low-dimensional coordinate based on a novel decentralized implementation of the Maximum Margin Matrix Factorization method, which optimizes the mapping between the network metrics and the level values. Simulation results for the round trip time, bandwidth, loss, and hop count metric confirm that HPM converges fast, is robust to parameter settings, scales well with increasing levels or system size, and adapts well to diverse metrics. A prototype deployment on

关键词： decentralized algorithm Service provision Application health monitoring Hierarchical decomposition K-means clustering decentralized matrix factorization

来源：评论

学校读者我要写书评

暂无评论

Integrated Electrical, Heating, and Water Distribution System to Accommodate Wind Power

引用

IEEE TRANSACTIONS ON SUSTAINABLE ENERGY 2021年第2期12卷 1100-1114页

作者： Li, Chunyan Yao, Yiming Xie, Kaigui Hu, Bo Niu, Tao Chongqing Univ Sch Elect Engn Chongqing 400044 Peoples R China Chongqing Univ Chongqing 400044 Peoples R China

With the rapid development of communication technology, the cooperation between electrical systems and other energy systems has become possible to promote wind power accommodation. As a common kind of energy carrier, the water distribution system has not been researched so far to help accommodate wind power when integrated with other energy systems. This paper proposes a novel model of integrated electrical, heating, and water distribution system to accommodate wind power. The optimal scheduling of heat pump and water pump are studied to make full use of the abundant wind power. The heat transmission loss is considered and thus the best siting of the heat pump can be determined. The demand response program is also included in the schedule structure to make the modeling more comprehensive. Moreover, a novel two-layer decentralized optimization algorithm is proposed to protect the data privacy of each energy system. The effectiveness of the proposed integrated system model and the two-layer decentralized algorithm are verified on two test systems. The results indicate that system integrations can reduce the wind power curtailment significantly and decrease the operation cost of each energy system. Besides, the proposed two-layer algorithm performs well in terms of the algorithm efficiency and scalability.

关键词： Wind power generation Water heating Heat pumps Resistance heating Cogeneration Mathematical model Wind power integrated energy system heating system water distribution system decentralized algorithm

来源：评论

学校读者我要写书评

暂无评论

DeEPCA: decentralized Exact PCA with Linear Convergence Rate

引用

JOURNAL OF MACHINE LEARNING RESEARCH 2021年第1期22卷 1-27页

作者： Ye, Haishan Zhang, Tong Xi An Jiao Tong Univ Sch Management Ctr Intelligent Decis Making & Machine Learning Xian Peoples R China Hong Kong Univ Sci & Technol Comp Sci & Math Kowloon Clear Water Bay Hong Kong Peoples R China

Due to the rapid growth of smart agents such as weakly connected computational nodes and sensors, developing decentralized algorithms that can perform computations on local agents becomes a major research direction. This paper considers the problem of decentralized principal components analysis (PCA), which is a statistical method widely used for data analysis. We introduce a technique called subspace tracking to reduce the communication cost, and apply it to power iterations. This leads to a decentralized PCA algorithm called DeEPCA, which has a convergence rate similar to that of the centralized PCA, while achieving the best communication complexity among existing decentralized PCA algorithms. DeEPCA is the first decentralized PCA algorithm with the number of communication rounds for each power iteration independent of target precision. Compared to existing algorithms, the proposed method is easier to tune in practice, with an improved overall communication cost. Our experiments validate the advantages of DeEPCA empirically.

关键词： decentralized algorithm Principal Component Analysis Subspace Tracking

来源：评论

学校读者我要写书评

暂无评论

Hybrid decentralized Data Analytics in Edge-Computing-Empowered IoT Networks

引用

IEEE INTERNET OF THINGS JOURNAL 2021年第9期8卷 7706-7716页

作者： Zhao, Liang Li, Fangyu Valero, Maria Kennesaw State Univ Dept Informat Technol Marietta GA 30060 USA Kennesaw State Univ Dept Elect & Comp Engn Marietta GA 30067 USA

Edge computing is emerging as a new infrastructure for Internet-of-Things (IoT) networks by placing computation and analytics near to where data are generated. This article presents a novel data analytics framework for edge computing. The framework is based on a new decentralized algorithm, which enables all the nodes to obtain the global optimal model without sharing raw data. The resulting scheme executes in a hybrid mode: local IoT nodes send computed information to edge nodes. The edge nodes cooperate with each other by exchanging analytics with their neighbors only. The presenting approach is analyzed and evaluated on various applications and the experimental results demonstrate the effectiveness of the proposed methodology in providing fast data analytics to edge computing infrastructure.

关键词： Data analytics decentralized algorithm edge computing Internet of Things (IoT)

来源：评论

学校读者我要写书评

暂无评论

Convergence Study of decentralized Min-Cost Subgraph algorithms for Multicast in Coded Networks

引用

IEEE TRANSACTIONS ON INFORMATION THEORY 2014年第1期60卷 410-421页

作者： Zhao, Fang Medard, Muriel Ozdaglar, Asuman Lun, Desmond MIT Alliance Res & Technol Future Urban Mobil Grp Singapore 138602 Singapore MIT Dept Elect Engn & Comp Sci Cambridge MA 02139 USA Rutgers State Univ Dept Comp Sci New Brunswick NJ 08901 USA

The problem of establishing minimum-cost multicast connections in coded networks can be viewed as an optimization problem, and decentralized algorithms were proposed by Lun et al. to compute the optimal subgraph using the dual subgradient method. However, the convergence rate problem for these algorithms remains open. There are limited results in the literature, which bound the amount of infeasibility of the primal solution recovered after each iterations or the convergence rate. However, due to the special structure of the network coding problem, we have an algorithm that generates a feasible solution after each iterations. In addition, the convergence rate of the primal problem is O(1/n) to a neighborhood of the optimal solution. We also propose heuristics to further improve our algorithm and demonstrate through simulations that the distributed algorithm converges to the optimal subgraph quickly and is robust against network topology changes.

关键词： Convergence rate decentralized algorithm multicast communication network coding subgradient method

来源：评论

学校读者我要写书评

暂无评论

Adaptive Kernel Learning in Heterogeneous Networks

引用

IEEE TRANSACTIONS ON SIGNAL AND INFORMATION PROCESSING OVER NETWORKS 2021年 7卷 423-437页

作者： Pradhan, Hrusikesha Bedi, Amrit Singh Koppel, Alec Rajawat, Ketan Indian Inst Technol Dept Engn Kanpur 208016 Uttar Pradesh India US Army Res Lab Adelphi MD 20783 USA

We consider learning in decentralized heterogeneous networks: agents seek to minimize a convex functional that aggregates data across the network, while only having access to their local data streams. We focus on the case where agents seek to estimate a regression function that belongs to a reproducing kernel Hilbert space (RKHS). To incentivize coordination while respecting network heterogeneity, we impose nonlinear proximity constraints. The resulting constrained stochastic optimization problem is solved using the functional variant of stochastic primal-dual (Arrow-Hurwicz) method which yields a decentralized algorithm. In order to avoid the model complexity from growing linearly over time, we project the primal iterates onto subspaces greedily constructed from kernel evaluations of agents' local observations. The resulting scheme, dubbed Heterogeneous Adaptive Learning with Kernels (HALK), allows us, for the first time, to characterize the precise trade-off between the optimality gap, constraint violation, and the model complexity. In particular, the proposed algorithm can be tuned to achieve zero constraint violation, an optimality gap of O(T-1/2 + alpha) after T iterations, where the number of elements retained in the dictionary is determined by 1/alpha. Simulations on a correlated spatio-temporal field estimation validate our theoretical results, which are corroborated in practice for networked oceanic sensing buoys estimating temperature and salinity from depth measurements.

关键词： Nonparametric regression decentralized algorithm statistical learning kernel methods stochastic optimization online learning stochastic primal-dual method

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：